Crédito — Risk Intelligence (mock)

[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_205":206,"_207":208},"routes/_app.x.$handle.$name",{"_97":96,"_32":45,"_105":87,"_27":106,"_13":14,"_107":-5,"_11":108,"_109":110,"_111":112,"_113":114,"_115":116,"_20":-5,"_117":-5,"_118":-5,"_119":25,"_120":121,"_122":59,"_123":-5,"_124":-5,"_125":126},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_43":44,"_32":45,"_38":46,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":57,"_58":59,"_60":-5,"_61":-5,"_62":63,"_64":65,"_66":-5,"_67":-5,"_68":-5,"_69":-5,"_70":71,"_72":71,"_73":74,"_75":74,"_76":74,"_77":71,"_78":71,"_79":80,"_81":82,"_83":84,"_85":25},"version",70,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":34,"_35":37,"_38":39,"_40":14,"_41":42,"_24":25},"previous",{"_30":31,"_32":28,"_33":34,"_35":36},"deleted",false,"deletedHere","path","app.js","latestPath","fileId","4e8d6832-9d76-11f1-be1c-1607ee4eb77e","name","fileType","script","content","/* Risk Intelligence — Crédito (mock F2.2): shell + Portafolio (Global · Concentración)\n Tokens solo vía CSS vars. Riesgo/deltas en NEUTRO; semáforo solo estado de límite. */\n(() => {\n const D = window.RI_DATA;\n\n // ---------- Rutas / navegación ----------\n const SECTIONS = [\n { id: \"vision-general\", label: \"Visión General\", ico: \"◎\", tabs: [], noTabsBar: true, placeholder: true },\n { id: \"portafolio\", label: \"Portafolio\", ico: \"▤\", tabs: [{ id: \"global\", label: \"Global\" }, { id: \"concentracion\", label: \"Concentración\" }] },\n { id: \"sensibilidades\", label: \"Sensibilidades\", ico: \"≋\", tabs: [{ id: \"supuestos\", label: \"Supuestos\" }], placeholder: true },\n { id: \"metricas-de-riesgo\", label: \"Métricas de Riesgo\", ico: \"∿\", tabs: [{ id: \"a-la-fecha\", label: \"A la Fecha\" }, { id: \"historico\", label: \"Histórico\" }], medida: true },\n { id: \"marginal\", label: \"Marginal\", ico: \"◮\", tabs: [{ id: \"global\", label: \"Global\" }], medida: true },\n { id: \"factores-de-riesgo\", label: \"Factores de Riesgo\", ico: \"❖\", tabs: [{ id: \"mapa-de-factores\", label: \"Mapa de Factores\" }] },\n { id: \"escenarios\", label: \"Escenarios\", ico: \"⇄\", tabs: [{ id: \"impacto-en-riesgo\", label: \"Impacto en Riesgo\" }], placeholder: true },\n { id: \"backtesting\", label: \"Backtesting\", ico: \"◔\", tabs: [{ id: \"resumen-excepciones\", label: \"Resumen de Excepciones\" }] },\n ];\n const S = {\n uiState: \"ok\", // ok | loading | empty | error | partial\n moneda: \"Todas\", // Todas | DOP | USD (filtro F1.10)\n donaVista: \"Calificación\",\n blVista: \"Calificación\",\n concTL: \"pct\", // pct | dop\n concTR: \"hhi\", // hhi | mayor\n expanded: {}, // rating → bool\n medida: \"VaR\", // VaR | ES | Std (solo Métricas de Riesgo y Marginal — cambia datos, nunca layout)\n btQ: \"1%\", // Backtesting BL: cuantil de diagnóstico (no cambia la medida oficial ni KPIs)\n mrDona: \"Emisor\", // vista dona A la Fecha: Emisor (default §8.1) | Tipo de instrumento\n mrBL: \"Calificación\", // dimensión divergentes Histórico\n mrExp: {}, // tabla A la Fecha: rating → bool\n mrExpH: {}, // tabla Histórico: rating → bool\n mgDona: \"Emisor\", // vista dona Marginal (default §10.1) | Calificación | Tipo de instrumento\n mgExp: {}, // tabla Marginal: rating → bool\n mgExpE: {}, // tabla Marginal: \"rating|emisor\" → bool (nivel instrumento)\n frExp: {}, // tabla Factores: factor → bool (drill factor → rating, §12)\n frRating: null, // selector Merton TR: null = predominante por share de VM (§7.1); persiste a través de estados (X-#11)\n filtersOpen: false,\n };\n let charts = [];\n\n function parseHash() {\n const h = location.hash.replace(/^#\\/?/, \"\");\n const [sec, tab] = h.split(\"/\");\n const section = SECTIONS.find(s => s.id === sec) || SECTIONS[0];\n const tabObj = section.tabs.find(t => t.id === tab) || section.tabs[0] || null;\n return { section, tab: tabObj };\n }\n function nav(sec, tab) { location.hash = \"#/\" + sec + (tab ? \"/\" + tab : \"\"); }\n\n // ---------- Formato (paridad producción: $13,476.90MM · M/MM nunca K) ----------\n const nf2 = new Intl.NumberFormat(\"en-US\", { minimumFractionDigits: 2, maximumFractionDigits: 2 });\n const nf1 = new Intl.NumberFormat(\"en-US\", { minimumFractionDigits: 1, maximumFractionDigits: 1 });\n const fMM = v => \"$\" + nf2.format(v) + \"MM\";\n const fPct = (v, d = 2) => (d === 1 ? nf1 : nf2).format(v) + \"%\";\n const fH = v => v.toFixed(4);\n const esc = s => String(s).replace(/&/g, \"&\").replace(//g, \">\").replace(/\"/g, \""\");\n\n // ---------- Agregación (DOP consolidado — L-4) ----------\n const dop = p => p.moneda === \"USD\" ? p.vm * D.TC : p.vm;\n const dopPrev = p => p.moneda === \"USD\" ? p.prev * D.TC_PREV : p.prev;\n const filt = () => D.positions.filter(p => S.moneda === \"Todas\" || p.moneda === S.moneda);\n function groupSum(list, key, val) {\n const m = new Map();\n list.forEach(p => m.set(p[key] ?? key(p), (m.get(p[key]) || 0) + val(p)));\n return m;\n }\n function aggBy(dim) { // dim: 'rating' | 'emisor' | 'tipo' | 'moneda'\n const m = new Map(), mp = new Map();\n filt().forEach(p => {\n const k = p[dim];\n m.set(k, (m.get(k) || 0) + dop(p));\n mp.set(k, (mp.get(k) || 0) + dopPrev(p));\n });\n return { cur: m, prev: mp };\n }\n const total = () => filt().reduce((a, p) => a + dop(p), 0);\n const totalPrev = () => filt().reduce((a, p) => a + dopPrev(p), 0);\n\n function emisorStats() {\n const { cur, prev } = aggBy(\"emisor\");\n const t = total();\n const rows = [...cur.entries()].map(([e, vm]) => {\n const monedas = new Set(filt().filter(p => p.emisor === e).map(p => p.moneda));\n const chip = monedas.size > 1 ? \"Mixta\" : [...monedas][0];\n const pct = t > 0 ? vm / t * 100 : 0;\n return { emisor: e, vm, prev: prev.get(e) || 0, pct, hhi: Math.pow(pct / 100, 2), chip };\n }).sort((a, b) => b.vm - a.vm);\n return rows;\n }\n function hhiOf(rows) { return rows.reduce((a, r) => a + r.hhi, 0); }\n function hhiPrev() {\n const { prev } = aggBy(\"emisor\");\n const tp = totalPrev();\n let h = 0; prev.forEach(v => { h += Math.pow(v / tp, 2); });\n return h;\n }\n function estadoEmisor(pct) { // semáforo SOLO estado de límite (con texto — nunca color solo)\n const lim = D.LIMITES.emisorPctInterno;\n if (pct > lim) return { cls: \"err\", txt: \"Exceso\" };\n if (pct >= lim * D.LIMITES.alertaFactor) return { cls: \"warn\", txt: \"Alerta\" };\n return { cls: \"ok\", txt: \"Ok\" };\n }\n\n // ---------- Paletas (mapping: ordinal azul; jamás semáforo en series) ----------\n const cs = getComputedStyle(document.documentElement);\n const TK = n => cs.getPropertyValue(n).trim();\n const ORDINAL = [\"#1f2a4e\", \"#28497c\", \"#0d77bf\", \"#4da3d8\", \"#8ec4e6\", \"#c4dff1\"]; // AAA→B (escala secuencial azul de colors-reports)\n const CATEG = [\"#0d77bf\", \"#1f2a4e\", \"#4da3d8\", \"#8ec4e6\", \"#6b7fae\", \"#a5b3d1\", \"#c4dff1\", \"#9aa4b8\"];\n const GRAY = \"#9aa4b8\";\n const ratColor = r => ORDINAL[D.RATINGS.indexOf(r)] ?? GRAY;\n\n // ---------- Render raíz ----------\n const $ = sel => document.querySelector(sel);\n function render() {\n charts.forEach(c => c.dispose()); charts = [];\n const { section, tab } = parseHash();\n renderSidebar(section);\n const main = $(\"#main\");\n const degraded = S.uiState === \"empty\" || S.uiState === \"error\";\n main.innerHTML = headHtml(section, tab, degraded) + tabsHtml(section, tab) + bodyHtml(section, tab);\n bindHead(); bindBody(section, tab);\n renderFilters();\n renderSwitch();\n }\n\n function renderSidebar(section) {\n $(\"#menu\").innerHTML = SECTIONS.map(s =>\n ``).join(\"\");\n document.querySelectorAll(\".menu-item\").forEach(b =>\n b.onclick = () => { const s = SECTIONS.find(x => x.id === b.dataset.sec); nav(s.id, s.tabs[0]?.id); });\n }\n\n // Cabecera de sección — badges FUERA de contenedores skeletonizables (X-2 / L-2)\n function headHtml(section, tab, degraded) {\n const sup = degraded\n ? `Supuesto: N/D`\n : `Supuesto: ${D.SUPUESTO.fuente} · ${D.SUPUESTO.metodo} · ${D.SUPUESTO.tc}`;\n const dataBadge = S.uiState === \"error\"\n ? `⚠ Error de datos`\n : S.uiState === \"empty\"\n ? `◌ Dataset no listo`\n : `Datos ${D.AS_OF} ⟳`;\n const crumb = section.noTabsBar\n ? `${section.label}`\n : `${section.label} / ${tab ? tab.label : \"\"}`;\n const selMedida = section.medida\n ? `${[\"VaR\", \"ES\", \"Std\"].map(m =>\n ``).join(\"\")}`\n : \"\";\n const chipVentana = section.id === \"backtesting\" && S.uiState !== \"empty\" && S.uiState !== \"error\"\n ? `Ventana: ${D.BT.cambios} cambio de supuesto`\n : \"\";\n return `

\n \n

\n ${selMedida}\n ⓘ ${sup}\n ${chipVentana}\n ${dataBadge}\n ${D.DATASET} ▾\n \n

`;\n }\n function tabsHtml(section, tab) {\n if (section.noTabsBar || !section.tabs.length) return `

`;\n return `

` + section.tabs.map(t =>\n ``).join(\"\") + `

`;\n }\n function bindHead() {\n const b = $(\"#btnFiltros\"); if (b) b.onclick = () => { S.filtersOpen = true; renderFilters(); };\n document.querySelectorAll(\"[data-med]\").forEach(x => x.onclick = () => { S.medida = x.dataset.med; render(); });\n document.querySelectorAll(\".tab-pill\").forEach(p =>\n p.onclick = () => { const { section } = parseHash(); nav(section.id, p.dataset.tab); });\n }\n\n // ---------- Cuerpos ----------\n function bodyHtml(section, tab) {\n if (section.placeholder) return placeholderHtml(section);\n if (section.id === \"portafolio\" && tab.id === \"global\") return globalHtml();\n if (section.id === \"portafolio\" && tab.id === \"concentracion\") return concHtml();\n if (section.id === \"metricas-de-riesgo\" && tab.id === \"a-la-fecha\") return mrFechaHtml();\n if (section.id === \"metricas-de-riesgo\" && tab.id === \"historico\") return mrHistHtml();\n if (section.id === \"marginal\") return mgHtml();\n if (section.id === \"factores-de-riesgo\") return frHtml();\n if (section.id === \"backtesting\") return btHtml();\n return \"\";\n }\n\n function placeholderHtml(section) {\n if (S.uiState === \"loading\") return `

`;\n if (S.uiState === \"error\") return `

${errBox()}

`;\n if (S.uiState === \"empty\") return `

${emptyBox()}

`;\n return `

\n ${section.label}\n Sección activa del módulo — contenido en la próxima iteración del prototipo.
\n El wireframe aprobado (03-wireframe-specs.md) define esta pantalla; este mock la incorpora incrementalmente.\n

`;\n }\n\n const emptyBox = () => `

◌\n Sin datos para este corte\n El dataset seleccionado aún no tiene resultados. Selecciona otro dataset en el selector superior o ejecuta el cálculo del corte.

`;\n const errBox = () => `

⚠\n No se pudo cargar la información\n Ocurrió un error al consultar los datos del corte. Puedes reintentar la carga.\n

`;\n\n function kpiCard(title, big, sub, rows) {\n return `

${title}

${big}

\n ${sub ? `

${sub}

` : \"\"}\n ${rows && rows.length ? `

` + rows.map(r =>\n `

${r[0]}${r[1]}

`).join(\"\") + `

` : \"\"}\n

`;\n }\n function kpiSkeletons() { return `

`.repeat(5); }\n function panelSkeletons(n = 4) { return `

`.repeat(n); }\n function statePanels(box, n = 4) { return Array(n).fill(`

${box}

`).join(\"\"); }\n\n function subtituloConsolidacion() {\n if (S.moneda === \"Todas\") return \"Consolidado en DOP · TC del corte\";\n return `Solo instrumentos ${S.moneda} · Consolidado en DOP · TC del corte`;\n }\n\n // ====== Portafolio › Global (§7.1) ======\n function globalHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisND(\"N/D\")}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisND(\"N/D\")}

${statePanels(errBox())}

`;\n\n const t = total(), tp = totalPrev(), dv = t - tp, dvp = tp > 0 ? dv / tp * 100 : 0;\n const pos = filt(), nPos = pos.length, nPosPrev = pos.filter(p => p.prev > 0).length;\n const emis = new Set(pos.map(p => p.emisor)).size;\n const emisPrev = new Set(pos.filter(p => p.prev > 0).map(p => p.emisor)).size;\n const byR = aggBy(\"rating\");\n let predom = null, predomPrev = null;\n byR.cur.forEach((v, k) => { if (!predom || v > byR.cur.get(predom)) predom = k; });\n byR.prev.forEach((v, k) => { if (!predomPrev || v > byR.prev.get(predomPrev)) predomPrev = k; });\n const shareNow = byR.cur.get(predom) / t * 100;\n const sharePrev = totalPrev() > 0 ? (byR.prev.get(predom) || 0) / totalPrev() * 100 : 0;\n const sameCat = predom === predomPrev;\n const arrow = sameCat ? (shareNow >= sharePrev ? \"▲\" : \"▼\") : \"\"; // flecha = Δ share del rating predominante actual (X-19)\n const eRows = emisorStats(); const hhi = hhiOf(eRows);\n const usoHHI = hhi / D.LIMITES.hhiInterno * 100;\n const excG = eRows.filter(r => estadoEmisor(r.pct).txt === \"Exceso\").length;\n const alG = eRows.filter(r => estadoEmisor(r.pct).txt === \"Alerta\").length;\n const worst = excG ? [\"err\", \"Exceso\"] : (alG || usoHHI >= 90) ? [\"warn\", \"Alerta\"] : usoHHI > 100 ? [\"err\", \"Exceso\"] : [\"ok\", \"Ok\"];\n const kpis =\n kpiCard(\"VM Total\", fMM(t), esc(subtituloConsolidacion()), [\n [\"anterior\", fMM(tp)], [\"Δ\", (dv >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dv)).slice(1), \"trend\"], [\"Δ%\", fPct(dvp), \"trend\"]]) +\n kpiCard(\"N° Posiciones\", String(nPos), null, [[\"anterior\", String(nPosPrev)], [\"Δ\", String(nPos - nPosPrev), \"trend\"]]) +\n kpiCard(\"N° Emisores\", String(emis), null, [[\"anterior\", String(emisPrev)], [\"Δ\", String(emis - emisPrev), \"trend\"]]) +\n kpiCard(\"Calificación predominante\", `${predom} ${arrow}`,\n `share ${fPct(shareNow, 1)} (ant. ${sameCat ? fPct(sharePrev, 1) : \"categoría distinta: \" + predomPrev})`, []) +\n kpiCard(\"Límite de concentración\", `${worst[1]}`,\n `HHI ${fH(hhi)} vs límite ${fH(D.LIMITES.hhiInterno)} · uso ${fPct(usoHHI, 1)}${excG || alG ? ` · límites por emisor: ${excG} incumplimiento(s), ${alG} alerta(s) — ver Concentración` : \" · límites por emisor sin excepciones\"}`, []);\n\n return `

${kpis}

Valor de Mercado por Calificación

Temporalidad: Año a la Fecha ▾ · Eje en DOP MM. Cada punto consolidado al TC de su fecha.

${S.uiState === \"partial\" ? errBox() : `\n

Composición del Portafolio

Participación sobre VM Total (DOP consolidado).${S.donaVista === \"Emisor\" ? \" Vista por emisor: top-7 + “Otros”.\" : \"\"}

Cambio en la Composición

Δ VM vs. corte anterior (DOP consolidado); barras en neutro — el signo lo comunica la posición y la etiqueta.

Tabla de Composición

${tablaGlobal()}

Variaciones al TC de cada corte; incluyen efecto FX. “—” = sin posiciones en el grupo; “N/D” = sin corte anterior (nunca 0).

`;\n }\n\n function kpisND(sym) {\n return [\"VM Total\", \"N° Posiciones\", \"N° Emisores\", \"Calificación predominante\", \"Límite de concentración\"]\n .map(t => kpiCard(t, sym, t === \"VM Total\" ? esc(subtituloConsolidacion()) : null, [])).join(\"\");\n }\n\n function tablaGlobal() {\n const t = total();\n const rows = D.RATINGS.map(r => {\n const ps = filt().filter(p => p.rating === r);\n if (!ps.length) return null;\n const agg = list => ({\n dop: list.filter(p => p.moneda === \"DOP\").reduce((a, p) => a + p.vm, 0),\n usd: list.filter(p => p.moneda === \"USD\").reduce((a, p) => a + p.vm, 0),\n tot: list.reduce((a, p) => a + dop(p), 0),\n prev: list.reduce((a, p) => a + dopPrev(p), 0),\n });\n const a = agg(ps);\n const emisores = [...new Set(ps.map(p => p.emisor))].map(e => {\n const l = ps.filter(p => p.emisor === e);\n const ea = agg(l);\n const monedas = new Set(l.map(p => p.moneda));\n return { e, a: ea, chip: monedas.size > 1 ? \"Mixta\" : [...monedas][0] };\n }).sort((x, y) => y.a.tot - x.a.tot);\n return { r, a, emisores };\n }).filter(Boolean);\n\n const cell = (v, unit) => v > 0 ? fMM(v) : \"—\";\n const varCells = a => {\n if (a.prev <= 0) return `N/DN/D`;\n const dv = a.tot - a.prev;\n return `${(dv >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dv)).slice(1)}${fPct(dv / a.prev * 100)}`;\n };\n let html = `\n \n `;\n rows.forEach(({ r, a, emisores }) => {\n const open = !!S.expanded[r];\n html += `\n \n ${varCells(a)}`;\n if (open) emisores.forEach(({ e, a: ea, chip }) => {\n html += `\n \n ${varCells(ea)}`;\n });\n });\n return html + `

Calificación	Valor de Mercado	% Global	Var (DOP)	Var %
DOP	USD	Total (DOP)
${open ? \"▾\" : \"▸\"}${r}	${cell(a.dop)}	${a.usd > 0 ? \"$\" + nf2.format(a.usd) + \"MM\" : \"—\"}	${fMM(a.tot)}	${fPct(a.tot / t * 100)}
${esc(e)} ${chip}	${cell(ea.dop)}	${ea.usd > 0 ? \"$\" + nf2.format(ea.usd) + \"MM\" : \"—\"}	${fMM(ea.tot)}	${fPct(ea.tot / t * 100)}

`;\n }\n\n // ====== Portafolio › Concentración (§7.2) ======\n function concHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDConc(\"N/D\")}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDConc(\"N/D\")}

${statePanels(errBox())}

`;\n\n const rows = emisorStats(); const hhi = hhiOf(rows); const hp = hhiPrev();\n const nEff = hhi > 0 ? 1 / hhi : 0;\n const top5 = rows.slice(0, 5).reduce((a, r) => a + r.pct, 0);\n const tp = totalPrev();\n const top5prev = rows.slice(0, 5).reduce((a, r) => a + (tp > 0 ? r.prev / tp * 100 : 0), 0);\n const mayor = rows[0];\n const estM = estadoEmisor(mayor.pct);\n const holg = D.LIMITES.emisorPctInterno - mayor.pct;\n const alertas = rows.filter(r => estadoEmisor(r.pct).txt === \"Alerta\").length;\n const exc = rows.filter(r => estadoEmisor(r.pct).txt === \"Exceso\").length;\n\n const kpis =\n kpiCard(\"HHI\", fH(hhi), `≈ ${nEff.toFixed(1)} emisores efectivos (1/HHI)`, [\n [\"anterior\", fH(hp)], [\"Δ\", (hhi - hp >= 0 ? \"+\" : \"−\") + fH(Math.abs(hhi - hp)), \"trend\"]]) +\n kpiCard(\"Share Top-5 emisores\", fPct(top5, 1), \"% del VM (DOP consolidado)\", [\n [\"anterior\", fPct(top5prev, 1)], [\"Δ\", fPct(top5 - top5prev, 1) + \" pp\", \"trend\"]]) +\n kpiCard(\"Mayor concentración\", esc(mayor.emisor), `${fPct(mayor.pct, 1)} del VM · ${holg >= 0 ? \"Holgura +\" + nf1.format(holg) : \"Exceso \" + nf1.format(Math.abs(holg))} pp vs. límite ${estM.txt}${S.moneda !== \"Todas\" ? \" · límites evaluados sobre el universo filtrado (\" + S.moneda + \"), no sobre el portafolio\" : \"\"}`, []) +\n kpiCard(\"Δ HHI vs. corte anterior\", `${(hhi - hp >= 0 ? \"+\" : \"−\") + fH(Math.abs(hhi - hp))}`, \"puntos de HHI\", []) +\n kpiCard(\"Estado de Límites\", `${exc ? \"Exceso\" : alertas ? \"Alerta\" : \"Ok\"}`,\n `Alertas: ${alertas} · Incumplimientos: ${exc} — límite interno ${fPct(D.LIMITES.emisorPctInterno, 1)} por emisor${S.moneda !== \"Todas\" ? \" · evaluados sobre el universo filtrado (\" + S.moneda + \")\" : \"\"}`, []);\n\n return `

${kpis}

Top Emisores

Top-7 + “Otros” por ${S.concTL === \"pct\" ? \"% del VM\" : \"VM (DOP consolidado)\"}.

${S.uiState === \"partial\" ? errBox() : `\n

Evolución de la Concentración

Línea punteada = límite interno. Cada punto consolidado al TC de su fecha.

Cambio en la Composición por Emisor

Δ VM vs. corte anterior — top-7 por |Δ| + “Otros” + Total; barras en neutro.

Tabla de Concentración por Emisor

${tablaConc(rows, hhi)}

Lista completa (exenta de top-7): un emisor en alerta o exceso jamás queda oculto. Límites de exposición sobre VM observado. Sin límite → “N/D”.

`;\n }\n function kpisNDConc(sym) {\n return [\"HHI\", \"Share Top-5 emisores\", \"Mayor concentración\", \"Δ HHI vs. corte anterior\", \"Estado de Límites\"]\n .map(t => kpiCard(t, sym, null, [])).join(\"\");\n }\n function tablaConc(rows, hhi) {\n let html = ``;\n rows.forEach(r => {\n const st = estadoEmisor(r.pct);\n const holg = D.LIMITES.emisorPctInterno - r.pct;\n html += `\n \n \n `;\n });\n return html + `

Emisor	Moneda	VM (DOP)	% VM	Aporte HHI	Límite interno	Holgura/Exceso
${esc(r.emisor)}	${r.chip}	${fMM(r.vm)}	${fPct(r.pct)}	${r.hhi.toFixed(4)}	${fPct(D.LIMITES.emisorPctInterno, 1)}	${(holg >= 0 ? \"+\" : \"\") + nf1.format(holg)} pp ${st.txt}

`;\n }\n\n // ====== Métricas de Riesgo (§8) ======\n // Las medidas se calculan SIEMPRE sobre la cartera consolidada completa — el filtro de\n // moneda NO aplica aquí (X-3/X-4: partición prohibida por cuotas; VaR no aditivo).\n const RM = () => D.RISK[S.medida];\n function vmFull(getVal) {\n const m = new Map();\n D.positions.forEach(p => m.set(p.rating, (m.get(p.rating) || 0) + getVal(p)));\n return m;\n }\n function mrAgg(prev = false) {\n const vm = vmFull(prev ? dopPrev : dop);\n const rows = D.RATINGS.filter(r => vm.has(r) && vm.get(r) > 0).map(r => {\n const stand = vm.get(r) * RM().rate[r];\n return { r, vm: vm.get(r), stand, contrib: stand * RM().divF[r] };\n });\n return {\n rows,\n total: rows.reduce((a, x) => a + x.contrib, 0),\n standSum: rows.reduce((a, x) => a + x.stand, 0),\n vmTot: rows.reduce((a, x) => a + x.vm, 0),\n };\n }\n function mrEmisores(prev = false) { // prorrateo por exposición dentro del rating (§2)\n const getVal = prev ? dopPrev : dop;\n const vm = vmFull(getVal);\n const m = new Map();\n D.positions.forEach(p => {\n if (!vm.get(p.rating)) return;\n const contribR = vm.get(p.rating) * RM().rate[p.rating] * RM().divF[p.rating];\n m.set(p.emisor, (m.get(p.emisor) || 0) + contribR * getVal(p) / vm.get(p.rating));\n });\n return [...m.entries()].sort((a, b) => b[1] - a[1]);\n }\n function mrTipos() { // contribución prorrateada por tipo de instrumento (misma base rating)\n const vm = vmFull(dop);\n const m = new Map();\n D.positions.forEach(p => {\n const contribR = vm.get(p.rating) * RM().rate[p.rating] * RM().divF[p.rating];\n m.set(p.tipo, (m.get(p.tipo) || 0) + contribR * dop(p) / vm.get(p.rating));\n });\n return [...m.entries()].sort((a, b) => b[1] - a[1]);\n }\n const mrMarkerIdx = () => D.DATES.indexOf(D.SUPUESTO_MARKERS[0].fecha);\n function mrHistSeries() { // ρ=20% antes del marcador → factor RHO_PREV_FACTOR (discontinuidad honesta)\n const mk = mrMarkerIdx();\n const cur = mrAgg();\n return D.RATINGS.map(r => ({\n r,\n data: D.DATES.map((_, i) => {\n if (i === D.DATES.length - 1) { const row = cur.rows.find(x => x.r === r); return row ? +row.contrib.toFixed(2) : 0; }\n const vmT = (D.HIST_SHARE[r][i] || 0) * D.HIST_TOTAL[i];\n return +(vmT * RM().rate[r] * RM().divF[r] * (i < mk ? D.RHO_PREV_FACTOR : 1)).toFixed(2);\n }),\n }));\n }\n function mrHistTotals() {\n const s = mrHistSeries();\n return D.DATES.map((_, i) => s.reduce((a, x) => a + x.data[i], 0));\n }\n const parseFecha = f => { const [d, m, y] = f.split(\"/\").map(Number); return new Date(y, m - 1, d); };\n function volAnualizada() { // σ descriptiva del VM del libro; anualiza por Δt promedio REAL (X-8)\n const vals = [...D.HIST_TOTAL]; vals[vals.length - 1] = mrAgg().vmTot;\n const rets = [];\n for (let i = 1; i < vals.length; i++) rets.push(Math.log(vals[i] / vals[i - 1]));\n const mu = rets.reduce((a, b) => a + b, 0) / rets.length;\n const sd = Math.sqrt(rets.reduce((a, b) => a + (b - mu) ** 2, 0) / (rets.length - 1));\n const spanDias = (parseFecha(D.DATES[D.DATES.length - 1]) - parseFecha(D.DATES[0])) / 86400000;\n const dtProm = spanDias / (D.DATES.length - 1);\n return { vol: sd * Math.sqrt(252 / dtProm) * 100, dtProm };\n }\n const LEYENDA_PRORRATEO = \"Contribución prorrateada por exposición — el modelo opera a nivel rating y no diferencia emisores dentro del rating (§2).\";\n\n function usoLimiteCard(total) {\n const lim = D.LIM_MEDIDA[S.medida];\n if (lim == null) return kpiCard(\"Uso Límite\",\n `N/D`,\n `Sin límite configurado para ${S.medida}`, []);\n const uso = total / lim * 100;\n const st = uso > 100 ? [\"err\", \"Incumplimiento\"] : uso >= 90 ? [\"warn\", \"Advertencia\"] : [\"ok\", \"Ok\"];\n return kpiCard(\"Uso Límite\", `${fPct(uso, 1)} ${st[1]}`,\n `${S.medida} vs. límite interno ${fMM(lim)} · cartera consolidada completa`, []);\n }\n const notaMonedaMR = `

Las medidas de riesgo se calculan sobre la cartera consolidada completa (DOP al TC del corte) y jamás se parten por moneda — el ${esc(S.medida)} no es aditivo entre sub-universos. El filtro de moneda no aplica en esta sección.

`;\n\n // --- Tab A la Fecha (§8.1) ---\n const MR_F_TITLES = () => [`${S.medida} Total`, `Razón ${S.medida}/VM`, \"Top Driver\", \"Beneficio de diversificación\", \"Uso Límite\"];\n function kpisNDMR(titles) { return titles.map(t => kpiCard(t, \"N/D\", null, [])).join(\"\"); }\n\n function mrFechaHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(MR_F_TITLES())}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(MR_F_TITLES())}

${statePanels(errBox())}

`;\n\n const a = mrAgg(), ap = mrAgg(true);\n const dv = a.total - ap.total, dvp = ap.total > 0 ? dv / ap.total * 100 : null;\n const razon = a.total / a.vmTot * 100;\n const emis = mrEmisores(), emisPrev = mrEmisores(true);\n const top = emis[0], topPrev = emisPrev[0];\n const ben = a.standSum - a.total, benPct = a.standSum > 0 ? ben / a.standSum * 100 : 0;\n const benTip = ben >= 0\n ? (S.medida === \"VaR\" ? \"Diversificación de cartera: Σ standalone − cartera. El VaR no es subaditivo — bajo concentración la brecha podría ser negativa.\" : `${S.medida} es subaditiva: el beneficio de diversificación es siempre ≥ 0.`)\n : \"El VaR no es subaditivo: penalización por concentración.\";\n const benLabel = ben >= 0 ? \"Beneficio de diversificación\" : \"Penalización por concentración\";\n\n const kpis =\n kpiCard(`${S.medida} Total`, fMM(a.total), \"Consolidado en DOP · TC del corte\", [\n [\"anterior\", fMM(ap.total)],\n [\"Δ\", (dv >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dv)).slice(1), \"trend\"],\n [\"Δ%\", dvp === null ? \"N/D\" : fPct(dvp), \"trend\"]]) +\n kpiCard(`Razón ${S.medida}/VM`, fPct(razon), \"sobre el VM consolidado de la cartera\", [\n [\"anterior\", ap.vmTot > 0 ? fPct(ap.total / ap.vmTot * 100) : \"N/D\"],\n [\"Δ\", ap.vmTot > 0 ? (razon - ap.total / ap.vmTot * 100 >= 0 ? \"+\" : \"−\") + nf2.format(Math.abs(razon - ap.total / ap.vmTot * 100)) + \" pp\" : \"N/D\", \"trend\"]]) +\n kpiCard(\"Top Driver\", esc(top[0]),\n `${fMM(top[1])} · ${fPct(top[1] / a.total * 100, 1)} del ${esc(S.medida)}
${esc(LEYENDA_PRORRATEO)}`, [\n [\"anterior\", esc(topPrev[0])]]) +\n kpiCard(`${benLabel} ⓘ`,\n (ben >= 0 ? \"\" : \"−\") + fMM(Math.abs(ben)),\n `${fPct(Math.abs(benPct), 1)} de la suma standalone (Σ ${esc(S.medida)} standalone − ${esc(S.medida)} de cartera)`, []) +\n usoLimiteCard(a.total);\n\n return `

${kpis}

${S.medida} por Calificación

Contribución a la cartera por rating, orden de escala AAA→B (DOP MM). Cero real = barra en 0; sin dato = barra ausente + “N/D”.

${S.uiState === \"partial\" ? errBox() : `\n

Contribución al ${S.medida}

${S.mrDona === \"Emisor\" ? \"Top-7 + “Otros”. \" : \"\"}${LEYENDA_PRORRATEO}

Standalone vs. Contribución por Calificación

La brecha entre ${S.medida} standalone y contribución a cartera es la diversificación. La vista por posición vive en Marginal.

Descomposición del ${S.medida}

${tablaMrFecha(a)}

Solo el VM se parte por moneda; el ${S.medida} va siempre en DOP consolidado (no aditivo entre monedas). Filas de emisor: ${LEYENDA_PRORRATEO} Las variaciones viven en el tab Histórico.

${notaMonedaMR}

`;\n }\n\n function tablaMrFecha(a) {\n let html = `\n \n `;\n D.RATINGS.forEach(r => {\n const ps = D.positions.filter(p => p.rating === r); if (!ps.length) return;\n const vmDop = ps.filter(p => p.moneda === \"DOP\").reduce((x, p) => x + p.vm, 0);\n const vmUsd = ps.filter(p => p.moneda === \"USD\").reduce((x, p) => x + p.vm, 0);\n const vmTot = ps.reduce((x, p) => x + dop(p), 0);\n const contribR = vmTot * RM().rate[r] * RM().divF[r];\n const open = !!S.mrExp[r];\n html += `\n \n `;\n if (open) {\n const byE = new Map();\n ps.forEach(p => byE.set(p.emisor, [...(byE.get(p.emisor) || []), p]));\n [...byE.entries()].sort((x, y) => y[1].reduce((s, p) => s + dop(p), 0) - x[1].reduce((s, p) => s + dop(p), 0)).forEach(([e, l]) => {\n const eDop = l.filter(p => p.moneda === \"DOP\").reduce((x, p) => x + p.vm, 0);\n const eUsd = l.filter(p => p.moneda === \"USD\").reduce((x, p) => x + p.vm, 0);\n const eTot = l.reduce((x, p) => x + dop(p), 0);\n const eContrib = contribR * eTot / vmTot;\n const chip = new Set(l.map(p => p.moneda)).size > 1 ? \"Mixta\" : l[0].moneda;\n html += `\n \n `;\n });\n }\n });\n return html + `

Calificación	Valor de Mercado	${S.medida} (DOP)	% del ${S.medida}
DOP	USD	Total (DOP)
${open ? \"▾\" : \"▸\"}${r}	${vmDop > 0 ? fMM(vmDop) : \"—\"}	${vmUsd > 0 ? \"$\" + nf2.format(vmUsd) + \"MM\" : \"—\"}	${fMM(vmTot)}	${fMM(contribR)}	${fPct(contribR / mrAgg().total * 100)}
${esc(e)} ${chip}	${eDop > 0 ? fMM(eDop) : \"—\"}	${eUsd > 0 ? \"$\" + nf2.format(eUsd) + \"MM\" : \"—\"}	${fMM(eTot)}	${fMM(eContrib)} prorr.	${fPct(eContrib / mrAgg().total * 100)}

`;\n }\n\n // --- Tab Histórico (§8.2) ---\n const MR_H_TITLES = () => [`${S.medida} Total`, `Rating de mayor Δ ${S.medida}`, \"Volatilidad anualizada del VM\", `Δ Razón ${S.medida}/VM`, \"Uso Límite\"];\n\n function mrHistHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(MR_H_TITLES())}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(MR_H_TITLES())}

${statePanels(errBox())}

`;\n\n const a = mrAgg(), ap = mrAgg(true);\n const dv = a.total - ap.total, dvp = ap.total > 0 ? dv / ap.total * 100 : null;\n // Rating de mayor Δ — desempate: |Δ|, luego orden de escala (X-14)\n const deltas = a.rows.map(x => {\n const prev = ap.rows.find(y => y.r === x.r);\n return { r: x.r, d: x.contrib - (prev ? prev.contrib : 0) };\n }).sort((x, y) => Math.abs(y.d) - Math.abs(x.d) || D.RATINGS.indexOf(x.r) - D.RATINGS.indexOf(y.r));\n const mx = deltas[0];\n const razon = a.total / a.vmTot * 100, razonPrev = ap.vmTot > 0 ? ap.total / ap.vmTot * 100 : null;\n const { vol, dtProm } = volAnualizada();\n\n const kpis =\n kpiCard(`${S.medida} Total`, fMM(a.total), \"Consolidado en DOP · TC del corte\", [\n [\"anterior\", fMM(ap.total)],\n [\"Δ\", (dv >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dv)).slice(1), \"trend\"],\n [\"Δ%\", dvp === null ? \"N/D\" : fPct(dvp), \"trend\"]]) +\n kpiCard(`Rating de mayor Δ ${S.medida}`, mx.r,\n `${(mx.d >= 0 ? \"+\" : \"−\") + fMM(Math.abs(mx.d))} vs. corte anterior · desempate por |Δ| y orden de escala`, [\n [\"anterior\", mx.r]]) +\n kpiCard(`Volatilidad anualizada del VM ⓘ`,\n fPct(vol, 1), \"no conmuta con el selector de medida\", []) +\n kpiCard(`Δ Razón ${S.medida}/VM`,\n razonPrev === null ? \"N/D\" : (razon - razonPrev >= 0 ? \"+\" : \"−\") + nf2.format(Math.abs(razon - razonPrev)) + \" pp\",\n \"separa efecto nivel de efecto tamaño del libro\", [\n [\"anterior\", razonPrev === null ? \"N/D\" : fPct(razonPrev)]]) +\n usoLimiteCard(a.total);\n\n return `

${kpis}

Evolución del ${S.medida} por Calificación

Cima del apilado = ${S.medida} Total. Línea vertical = cambio de supuesto — la serie no se lee como continua entre supuestos distintos (§2). Cada punto consolidado al TC de su fecha.

${S.uiState === \"partial\" ? errBox() : `\n

Evolución de la Razón ${S.medida}/VM

Intensidad de riesgo del libro (%). Mismos marcadores de cambio de supuesto que la serie de la izquierda.

Cambio del ${S.medida} vs. Corte Anterior

Δ vs. corte anterior (28/01 — mismo supuesto que el as_of; los cortes previos al 03/01 no son comparables punto a punto por cambio de supuesto). Barras en neutro.${S.mrBL === \"Emisor\" ? \" Top-7 por |Δ| + “Otros”. \" + LEYENDA_PRORRATEO : \"\"}

Comparativa por Calificación

${tablaMrHist(a, ap)}

Variaciones al TC de cada corte; incluyen efecto FX. Cada punto consolidado al TC de su fecha. Δ solo entre cortes con el mismo supuesto — supuesto distinto → “N/D”, nunca 0. Filas de emisor: ${LEYENDA_PRORRATEO}

${notaMonedaMR}

`;\n }\n\n function tablaMrHist(a, ap) {\n let html = `\n \n `;\n D.RATINGS.forEach(r => {\n const ps = D.positions.filter(p => p.rating === r); if (!ps.length) return;\n const vmDop = ps.filter(p => p.moneda === \"DOP\").reduce((x, p) => x + p.vm, 0);\n const vmUsd = ps.filter(p => p.moneda === \"USD\").reduce((x, p) => x + p.vm, 0);\n const vmTot = ps.reduce((x, p) => x + dop(p), 0);\n const cur = a.rows.find(x => x.r === r), prev = ap.rows.find(x => x.r === r);\n const varCells = prev\n ? ``\n : ``;\n const open = !!S.mrExpH[r];\n html += `\n \n ${varCells}`;\n if (open) {\n const byE = new Map();\n ps.forEach(p => byE.set(p.emisor, [...(byE.get(p.emisor) || []), p]));\n [...byE.entries()].forEach(([e, l]) => {\n const eTot = l.reduce((x, p) => x + dop(p), 0), ePrev = l.reduce((x, p) => x + dopPrev(p), 0);\n const eContrib = cur.contrib * eTot / vmTot;\n const eContribPrev = prev && ap.rows.find(x => x.r === r) ? prev.contrib * ePrev / ap.rows.find(x => x.r === r).vm : null;\n const eDop = l.filter(p => p.moneda === \"DOP\").reduce((x, p) => x + p.vm, 0);\n const eUsd = l.filter(p => p.moneda === \"USD\").reduce((x, p) => x + p.vm, 0);\n const chip = new Set(l.map(p => p.moneda)).size > 1 ? \"Mixta\" : l[0].moneda;\n const vCells = eContribPrev && ePrev > 0\n ? ``\n : ``;\n html += `\n \n ${vCells}`;\n });\n }\n });\n return html + `

Calificación	Valor de Mercado	${S.medida} (DOP)	Var (DOP)	Var %
DOP	USD	Total (DOP)
${(cur.contrib - prev.contrib >= 0 ? \"+\" : \"−\") + fMM(Math.abs(cur.contrib - prev.contrib)).slice(1)}	${fPct((cur.contrib - prev.contrib) / prev.contrib * 100)}	N/D	N/D
${open ? \"▾\" : \"▸\"}${r}	${vmDop > 0 ? fMM(vmDop) : \"—\"}	${vmUsd > 0 ? \"$\" + nf2.format(vmUsd) + \"MM\" : \"—\"}	${fMM(vmTot)}	${fMM(cur.contrib)}
${(eContrib - eContribPrev >= 0 ? \"+\" : \"−\") + fMM(Math.abs(eContrib - eContribPrev)).slice(1)}	${fPct((eContrib - eContribPrev) / eContribPrev * 100)}	N/D	N/D
${esc(e)} ${chip}	${eDop > 0 ? fMM(eDop) : \"—\"}	${eUsd > 0 ? \"$\" + nf2.format(eUsd) + \"MM\" : \"—\"}	${fMM(eTot)}	${fMM(eContrib)} prorr.

`;\n }\n\n // ====== Marginal (§10) — salidas directas por posición: leave-one-out (marginal),\n // Euler (componental) y standalone. La cartera es SIEMPRE la consolidada completa.\n const NOTA_GRANULARIDAD = \"En el modelo, los emisores de una misma calificación comparten distribución — las diferencias por posición provienen de su exposición y de la realización simulada de su escenario.\";\n const MG_TITLES = () => [`${S.medida} Marginal promedio`, \"Mayor contribución\", \"Posiciones sobre umbral\", \"HHI de contribuciones\", \"Uso Límite\"];\n function mgRows(prev = false) {\n const getVal = prev ? dopPrev : dop;\n // base: standalone y componental Euler por posición (misma fórmula que Métricas por rating)\n const base = D.positions.map(p => {\n const v = getVal(p);\n const stand = v * RM().rate[p.rating];\n return { p, v, stand, comp: stand * RM().divF[p.rating] };\n });\n // diversificadoras (realización simulada — §2: no es atributo del emisor): componental\n // negativa; el excedente se redistribuye proporcionalmente INTRA-RATING → los subtotales\n // por rating y el total coinciden EXACTO con la pantalla Métricas de Riesgo.\n D.RATINGS.forEach(r => {\n const inR = base.filter(x => x.p.rating === r);\n const divs = inR.filter(x => x.p.div), rest = inR.filter(x => !x.p.div);\n if (!divs.length || !rest.length) return;\n let excess = 0;\n divs.forEach(x => { const nc = -0.10 * x.stand; excess += x.comp - nc; x.comp = nc; });\n const restSum = rest.reduce((a, x) => a + x.comp, 0);\n rest.forEach(x => { x.comp += excess * x.comp / restSum; });\n });\n base.forEach(x => {\n x.marg = x.comp * x.p.mgF;\n x.margPct = x.v > 0 ? x.marg / x.v * 100 : 0;\n });\n return base;\n }\n const fSg = v => (v < 0 ? \"−\" : \"\") + fMM(Math.abs(v));\n\n function mgHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(MG_TITLES())}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(MG_TITLES())}

${statePanels(errBox())}

`;\n\n const rows = mgRows();\n const total = rows.reduce((a, x) => a + x.comp, 0); // Σ componental = total de Métricas (Euler)\n const margSum = rows.reduce((a, x) => a + x.marg, 0);\n const mean = margSum / rows.length;\n const meanPrev = mgRows(true).reduce((a, x) => a + x.marg, 0) / rows.length;\n const dMean = mean - meanPrev, dMeanP = meanPrev !== 0 ? dMean / Math.abs(meanPrev) * 100 : null;\n const top = [...rows].sort((a, b) => b.comp - a.comp)[0];\n const lims = D.MG_LIMITS[S.medida];\n const nSobre = lims.hyp != null ? rows.filter(x => x.marg > lims.hyp).length : null;\n const pos = rows.filter(x => x.comp > 0);\n const posSum = pos.reduce((a, x) => a + x.comp, 0);\n const hhi = pos.reduce((a, x) => a + (x.comp / posSum) ** 2, 0);\n const negs = rows.filter(x => x.comp < 0);\n const negNet = negs.reduce((a, x) => a + x.comp, 0);\n\n const kpis =\n kpiCard(`${S.medida} Marginal promedio ⓘ`,\n fSg(mean), subtituloConsolidacion(), [\n [\"anterior\", fSg(meanPrev)],\n [\"Δ\", (dMean >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dMean)).slice(1)],\n [\"Δ%\", dMeanP === null ? \"N/D\" : (dMeanP >= 0 ? \"+\" : \"\") + nf1.format(dMeanP) + \"%\"]]) +\n kpiCard(`Mayor contribución ⓘ`,\n fMM(top.comp),\n `${esc(top.p.inst)} · ${esc(top.p.emisor)}
${fPct(top.comp / total * 100, 1)} del ${esc(S.medida)} (base componental — aditiva)`, []) +\n kpiCard(`Posiciones sobre umbral ⓘ`,\n nSobre === null ? \"N/D\" : String(nSobre),\n nSobre === null ? \"Límite de riesgo absoluto: N/D\" : `Marginal > ${fMM(lims.hyp)} (hipérbola iso-risk)`, []) +\n kpiCard(`HHI de contribuciones ⓘ`,\n fH(hhi), `≈ ${nf0(1 / hhi)} posiciones efectivas (1/HHI) · sobre contribuciones positivas`, []) +\n usoLimiteCard(total);\n\n const glos = `

ⓘ Definiciones — Marginal: riesgo(cartera) − riesgo(cartera sin la posición), leave-one-out; no aditiva (en este corte Σ marginales = ${fMM(margSum)} ≠ ${esc(S.medida)} total ${fMM(total)}). Contribución (componental): descomposición de Euler — Σ componentales = ${esc(S.medida)} total exacto. Standalone: riesgo de la posición en aislamiento; standalone − componental = beneficio de diversificación.

`;\n\n return `${glos}

${kpis}

Iso-risk: ${S.medida} marginal vs. tamaño

Cada punto = instrumento; X·Y = marginal en DOP. Overlays de límite conmutan con la medida; sin límite configurado → “Límite: N/D”. Diversificadoras bajo Y=0. ${NOTA_GRANULARIDAD}

${S.uiState === \"partial\" ? errBox() : `\n

Contribución (componental)

${S.mgDona === \"Emisor\" ? \"Top-7 + “Otros”. \" : \"\"}Solo contribuciones positivas — una dona no representa negativos. Diversificadoras: ${negs.length} posiciones, neto −${fMM(Math.abs(negNet)).slice(1)} (fuera de la dona).

Ranking por ${S.medida} marginal

Top-7 posiciones por |marginal| mostrando signo: añade (+) / diversifica (−). “Otros (+)” y “Otros (−)” separados — un neto ocultaría magnitudes de ambos signos.

Detalle por posición

${tablaMg(rows, total)}

Lista completa (sin colapso top-7). Marginal solo a nivel instrumento — no es aditiva. Los % pueden superar 100% por contribuciones negativas. Var refiere al Marginal; variaciones al TC de cada corte (incluyen efecto FX). Solo el VM se consolida; el ${S.medida} jamás se parte por moneda. ${NOTA_GRANULARIDAD}

${notaMonedaMR}

`;\n }\n\n function tablaMg(rows, total) {\n const prevRows = mgRows(true);\n let html = `\n \n \n \n \n `;\n D.RATINGS.forEach(r => {\n const inR = rows.filter(x => x.p.rating === r); if (!inR.length) return;\n const vmR = inR.reduce((a, x) => a + x.v, 0), compR = inR.reduce((a, x) => a + x.comp, 0);\n const standR = inR.reduce((a, x) => a + x.stand, 0);\n const open = !!S.mgExp[r];\n html += `\n \n \n `;\n if (!open) return;\n const byE = new Map();\n inR.forEach(x => byE.set(x.p.emisor, [...(byE.get(x.p.emisor) || []), x]));\n [...byE.entries()].sort((a, b) => b[1].reduce((s, x) => s + x.v, 0) - a[1].reduce((s, x) => s + x.v, 0)).forEach(([e, l]) => {\n const key = r + \"|\" + e, openE = !!S.mgExpE[key];\n const vmE = l.reduce((a, x) => a + x.v, 0), compE = l.reduce((a, x) => a + x.comp, 0), standE = l.reduce((a, x) => a + x.stand, 0);\n const chip = new Set(l.map(x => x.p.moneda)).size > 1 ? \"Mixta\" : l[0].p.moneda;\n html += `\n \n \n `;\n if (!openE) return;\n l.sort((a, b) => b.v - a.v).forEach(x => {\n const prev = prevRows.find(y => y.p.inst === x.p.inst && y.p.emisor === x.p.emisor);\n const dv = prev ? x.marg - prev.marg : null;\n const dvp = prev && prev.marg !== 0 ? dv / Math.abs(prev.marg) * 100 : null;\n html += `\n \n \n \n \n `;\n });\n });\n });\n return html + `

Calificación / Emisor / Instrumento	Moneda	VM (DOP)	${S.medida} Marginal	Componental	% del total	Standalone	Benef. diversif.	Var Marginal	Var %
${open ? \"▾\" : \"▸\"}${r}	—	${fMM(vmR)}	—	${fSg(compR)}	${fPct(compR / total * 100)}	${fMM(standR)}	${fSg(standR - compR)}	—	—
${openE ? \"▾\" : \"▸\"}${esc(e)}	${chip}	${fMM(vmE)}	—	${fSg(compE)}	${fPct(compE / total * 100)}	${fMM(standE)}	${fSg(standE - compE)}	—	—
${esc(x.p.inst)}	${x.p.moneda}	${fMM(x.v)}	${fSg(x.marg)}	${fSg(x.comp)}	${fPct(x.comp / total * 100)}	${fMM(x.stand)}	${fSg(x.stand - x.comp)}	${dv === null ? \"N/D\" : (dv >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dv)).slice(1)}	${dvp === null ? \"N/D\" : (dvp >= 0 ? \"+\" : \"\") + nf1.format(dvp) + \"%\"}

`;\n }\n\n function chartsMarginal() {\n const rows = mgRows();\n const lims = D.MG_LIMITS[S.medida];\n // TL — scatter iso-risk. Series por rating (paleta ordinal, jamás semáforo) + 3 overlays\n // con entrada de leyenda SIEMPRE presente (layout estable ante el selector — N/D si null).\n const maxX = Math.max(...rows.map(x => x.v)) * 1.12;\n const scatterSeries = D.RATINGS.filter(r => rows.some(x => x.p.rating === r)).map(r => ({\n name: r, type: \"scatter\", symbolSize: 13, itemStyle: { color: ratColor(r), opacity: 0.7 },\n data: rows.filter(x => x.p.rating === r).map(x => ({ value: [+x.v.toFixed(2), +x.margPct.toFixed(3)], meta: x })),\n }));\n const maxY = Math.max(...rows.map(x => x.margPct)) * 1.45;\n const minY = Math.min(0, ...rows.map(x => x.margPct)) * 2.2 - 0.4;\n const hypPts = lims.hyp != null\n ? Array.from({ length: 240 }, (_, i) => { const x = maxX * (i + 1) / 240; const y = 100 * lims.hyp / x; return y <= maxY ? [x, y] : null; }).filter(Boolean)\n : [];\n const nmH = `Límite %-riesgo: ${lims.h != null ? fPct(lims.h, 1) : \"N/D\"}`;\n const nmV = `Límite de tamaño: ${lims.v != null ? fMM(lims.v) : \"N/D\"}`;\n const nmY = `Iso-risk (por posición): ${lims.hyp != null ? fMM(lims.hyp) : \"N/D\"}`;\n mkChart(\"chMG_TL\", {\n grid: { left: 52, right: 14, top: 34, bottom: 74 },\n legend: { bottom: 0, itemWidth: 10, itemHeight: 10, textStyle: { fontSize: 9 } },\n tooltip: { trigger: \"item\", formatter: p => {\n const m = p.data && p.data.meta; if (!m) return p.seriesName;\n return `${esc(m.p.inst)} · ${esc(m.p.emisor)}
Moneda: ${m.p.moneda} · Rating actual: ${m.p.rating}
VM: ${fMM(m.v)}
Marginal: ${fPct(m.margPct)} · ${fSg(m.marg)}
${esc(NOTA_GRANULARIDAD)}`;\n } },\n xAxis: { name: \"VM (DOP MM)\", nameLocation: \"middle\", nameGap: 22, nameTextStyle: { fontSize: 9 }, type: \"value\", max: +maxX.toFixed(0), axisLabel: { fontSize: 9, formatter: v => \"$\" + nf0(v) } },\n yAxis: { name: `${S.medida} marginal / VM (%)`, nameTextStyle: { fontSize: 9, align: \"left\" }, nameGap: 10, type: \"value\", max: +maxY.toFixed(1), min: +minY.toFixed(1), axisLabel: { fontSize: 9, formatter: v => nf1.format(v) + \"%\" } },\n dataZoom: [{ type: \"inside\", xAxisIndex: 0 }, { type: \"inside\", yAxisIndex: 0 }],\n graphic: [\n { type: \"text\", right: 28, top: 40, style: { text: \"C · mayor riesgo absoluto\\n«fallen angels viven aquí»\", fontSize: 9, fill: TK(\"--text-2\"), fontWeight: \"bold\" } },\n { type: \"text\", left: 62, top: 40, style: { text: \"B · mayor % de riesgo\", fontSize: 9, fill: TK(\"--text-2\") } },\n { type: \"text\", right: 28, bottom: 108, style: { text: \"A · mayor tamaño absoluto\", fontSize: 9, fill: TK(\"--text-2\") } },\n ],\n series: [\n ...scatterSeries,\n { name: nmH, type: \"line\", symbol: \"none\", lineStyle: { type: \"dashed\", width: 1.4, color: TK(\"--secondary\") }, color: TK(\"--secondary\"),\n data: lims.h != null ? [[0, lims.h], [maxX, lims.h]] : [],\n markLine: { silent: true, symbol: \"none\", lineStyle: { color: TK(\"--text-2\"), width: 1 }, label: { show: false }, data: [{ yAxis: 0 }] } },\n { name: nmV, type: \"line\", symbol: \"none\", lineStyle: { type: \"dashed\", width: 1.4, color: TK(\"--primary\") }, color: TK(\"--primary\"),\n data: lims.v != null ? [[lims.v, minY], [lims.v, maxY]] : [] },\n { name: nmY, type: \"line\", symbol: \"none\", lineStyle: { width: 2, type: \"dashed\", color: TK(\"--sem-warn\") }, color: TK(\"--sem-warn\"),\n data: hypPts, tooltip: { show: false } },\n ],\n });\n // TR — dona componental (solo positivas)\n if (S.uiState !== \"partial\") {\n const pos = rows.filter(x => x.comp > 0);\n let entries;\n if (S.mgDona === \"Emisor\") {\n const m = new Map(); pos.forEach(x => m.set(x.p.emisor, (m.get(x.p.emisor) || 0) + x.comp));\n entries = [...m.entries()].sort((a, b) => b[1] - a[1]);\n if (entries.length > 8) { const t7 = entries.slice(0, 7), rest = entries.slice(7).reduce((a, e) => a + e[1], 0); entries = [...t7, [\"Otros\", rest]]; }\n } else if (S.mgDona === \"Calificación\") {\n const m = new Map(); pos.forEach(x => m.set(x.p.rating, (m.get(x.p.rating) || 0) + x.comp));\n entries = D.RATINGS.filter(r => m.has(r)).map(r => [r, m.get(r)]);\n } else {\n const m = new Map(); pos.forEach(x => m.set(x.p.tipo, (m.get(x.p.tipo) || 0) + x.comp));\n entries = [...m.entries()].sort((a, b) => b[1] - a[1]);\n }\n const colors = S.mgDona === \"Calificación\" ? entries.map(e => ratColor(e[0]))\n : entries.map((e, i) => e[0] === \"Otros\" ? GRAY : CATEG[i % CATEG.length]);\n mkChart(\"chMG_TR\", {\n tooltip: { trigger: \"item\", valueFormatter: v => fMM(v) },\n legend: { orient: \"vertical\", right: 0, top: \"middle\", textStyle: { fontSize: 10 }, formatter: n => { const e = entries.find(x => x[0] === n); return `${n} ${fMM(e[1])}`; } },\n series: [{ type: \"pie\", radius: [\"52%\", \"74%\"], center: [\"32%\", \"50%\"],\n label: { formatter: p => fPct(p.percent, 1), fontSize: 10 },\n data: entries.map((e, i) => ({ name: e[0], value: +e[1].toFixed(2), itemStyle: { color: colors[i] } })) }],\n });\n }\n // BL — divergentes top-7 |marginal| + Otros(+)/Otros(−) separados (neutro, signo por posición)\n const sorted = [...rows].sort((a, b) => Math.abs(b.marg) - Math.abs(a.marg));\n const t7 = sorted.slice(0, 7), rest = sorted.slice(7);\n const oPos = rest.filter(x => x.marg >= 0).reduce((a, x) => a + x.marg, 0);\n const oNeg = rest.filter(x => x.marg < 0).reduce((a, x) => a + x.marg, 0);\n const items = [\n ...t7.map(x => ({ k: x.p.inst, v: x.marg })),\n { k: \"Otros (+)\", v: oPos }, { k: \"Otros (−)\", v: oNeg },\n ];\n mkChart(\"chMG_BL\", {\n grid: { left: 56, right: 16, top: 24, bottom: 66 },\n tooltip: { trigger: \"axis\", valueFormatter: v => (v >= 0 ? \"+\" : \"−\") + fMM(Math.abs(v)).slice(1) },\n xAxis: { type: \"category\", data: items.map(i => i.k), axisLabel: { fontSize: 8.5, rotate: 36 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => (v < 0 ? \"−\" : \"\") + \"$\" + nf0(Math.abs(v)) + \"MM\", fontSize: 10 } },\n series: [{ type: \"bar\", barWidth: 20,\n itemStyle: { color: p => items[p.dataIndex].k.startsWith(\"Otros\") ? GRAY : TK(\"--primary\") },\n label: { show: true, position: \"top\", fontSize: 9, color: TK(\"--text-2\"),\n formatter: p => items[p.dataIndex].v === 0 ? \"sin posiciones\" : (items[p.dataIndex].v >= 0 ? \"+\" : \"−\") + nf1.format(Math.abs(items[p.dataIndex].v)) },\n data: items.map(i => +i.v.toFixed(2)) }],\n });\n }\n\n // ---------- Charts ----------\n function mkChart(id, opt) {\n const el = document.getElementById(id);\n if (!el) return;\n const c = echarts.init(el, null, { renderer: \"canvas\" });\n c.setOption(opt); charts.push(c);\n }\n const tipMM = p => (Array.isArray(p) ? p : [p]).map(x => `${x.marker} ${x.seriesName || x.name}: ${fMM(x.value)}`).join(\"
\");\n\n function chartsGlobal() {\n // TL — área apilada VM por rating (paleta ordinal azul, jamás semáforo)\n const shares = D.HIST_SHARE, totals = [...D.HIST_TOTAL];\n const byR = aggBy(\"rating\"); const t = total();\n totals[totals.length - 1] = t;\n const series = D.RATINGS.map(r => ({\n name: r, type: \"line\", stack: \"vm\", areaStyle: { opacity: 0.85 }, symbol: \"none\",\n lineStyle: { width: 1 }, color: ratColor(r),\n data: D.DATES.map((_, i) => i === D.DATES.length - 1\n ? +(byR.cur.get(r) || 0).toFixed(2)\n : +((shares[r][i] || 0) * totals[i]).toFixed(2)),\n }));\n mkChart(\"chTL\", {\n grid: { left: 60, right: 16, top: 24, bottom: 44 },\n tooltip: { trigger: \"axis\", valueFormatter: v => fMM(v) },\n legend: { bottom: 0, itemWidth: 10, itemHeight: 10, textStyle: { fontSize: 10 } },\n xAxis: { type: \"category\", data: D.DATES, axisLabel: { fontSize: 9, rotate: 40 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n series,\n });\n // TR — dona con \"Vista por\"\n if (S.uiState !== \"partial\") {\n const dim = { \"Calificación\": \"rating\", \"Emisor\": \"emisor\", \"Tipo de instrumento\": \"tipo\", \"Moneda\": \"moneda\" }[S.donaVista];\n let entries = [...aggBy(dim).cur.entries()].sort((a, b) => b[1] - a[1]);\n if (entries.length > 8) { // máx 8 categorías: top-7 + Otros (L-10)\n const top = entries.slice(0, 7), rest = entries.slice(7).reduce((a, e) => a + e[1], 0);\n entries = [...top, [\"Otros\", rest]];\n }\n const colors = dim === \"rating\" ? entries.map(e => e[0] === \"Otros\" ? GRAY : ratColor(e[0]))\n : entries.map((e, i) => e[0] === \"Otros\" ? GRAY : CATEG[i % CATEG.length]);\n mkChart(\"chTR\", {\n tooltip: { trigger: \"item\", valueFormatter: v => fMM(v) },\n legend: { orient: \"vertical\", right: 0, top: \"middle\", textStyle: { fontSize: 10 }, formatter: n => { const e = entries.find(x => x[0] === n); return `${n} ${fMM(e[1])}`; } },\n series: [{ type: \"pie\", radius: [\"52%\", \"74%\"], center: [\"32%\", \"50%\"],\n label: { formatter: p => fPct(p.percent, 1), fontSize: 10 },\n data: entries.map((e, i) => ({ name: e[0], value: +e[1].toFixed(2), itemStyle: { color: colors[i] } })) }],\n });\n }\n // BL — divergentes Δ (NEUTRO — signo por posición+etiqueta, X-3)\n const dimB = { \"Calificación\": \"rating\", \"Emisor\": \"emisor\", \"Tipo de instrumento\": \"tipo\" }[S.blVista];\n const ag = aggBy(dimB);\n let items = [...ag.cur.keys()].map(k => {\n const c = ag.cur.get(k) || 0, p = ag.prev.get(k) || 0;\n return { k, d: c - p, dp: p > 0 ? (c - p) / p * 100 : null };\n });\n if (dimB === \"emisor\" && items.length > 7) {\n items.sort((a, b) => Math.abs(b.d) - Math.abs(a.d));\n const top = items.slice(0, 7), rest = items.slice(7);\n items = [...top, { k: \"Otros\", d: rest.reduce((a, x) => a + x.d, 0), dp: null }];\n } else if (dimB === \"rating\") {\n items.sort((a, b) => D.RATINGS.indexOf(a.k) - D.RATINGS.indexOf(b.k));\n }\n items.push({ k: \"Total\", d: total() - totalPrev(), dp: totalPrev() > 0 ? (total() - totalPrev()) / totalPrev() * 100 : null });\n mkChart(\"chBL\", {\n grid: { left: 60, right: 16, top: 28, bottom: 54 },\n tooltip: { trigger: \"axis\", valueFormatter: v => (v >= 0 ? \"+\" : \"−\") + fMM(Math.abs(v)).slice(1) },\n xAxis: { type: \"category\", data: items.map(i => i.k), axisLabel: { fontSize: 9, rotate: 32 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n series: [{ type: \"bar\", barWidth: 22,\n itemStyle: { color: p => p.dataIndex === items.length - 1 ? TK(\"--secondary\") : TK(\"--primary\") }, // neutro siempre\n label: { show: true, position: \"top\", fontSize: 9, color: TK(\"--text-2\"),\n formatter: p => items[p.dataIndex].dp === null ? \"\" : (items[p.dataIndex].dp >= 0 ? \"+\" : \"\") + nf1.format(items[p.dataIndex].dp) + \"%\" },\n data: items.map(i => +i.d.toFixed(2)) }],\n });\n }\n\n function chartsConc() {\n const rows = emisorStats(); const t = total();\n // TL — top-7 + Otros barras horizontales\n let top = rows.slice(0, 7);\n const rest = rows.slice(7);\n const data = [...top.map(r => ({ name: r.emisor, v: S.concTL === \"pct\" ? r.pct : r.vm }))];\n if (rest.length) data.push({ name: \"Otros\", v: rest.reduce((a, r) => a + (S.concTL === \"pct\" ? r.pct : r.vm), 0) });\n data.reverse();\n mkChart(\"chCTL\", {\n grid: { left: 130, right: 50, top: 8, bottom: 24 },\n tooltip: { trigger: \"axis\", valueFormatter: v => S.concTL === \"pct\" ? fPct(v) : fMM(v) },\n xAxis: { type: \"value\", axisLabel: { formatter: v => S.concTL === \"pct\" ? v + \"%\" : \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n yAxis: { type: \"category\", data: data.map(d => d.name), axisLabel: { fontSize: 10 } },\n series: [{ type: \"bar\", barWidth: 14,\n itemStyle: { color: p => data[p.dataIndex].name === \"Otros\" ? GRAY : TK(\"--primary\") },\n label: { show: true, position: \"right\", fontSize: 9.5, formatter: p => S.concTL === \"pct\" ? fPct(p.value, 1) : fMM(p.value) },\n data: data.map(d => +d.v.toFixed(2)) }],\n });\n // TR — evolución HHI / mayor% con markLine límite\n if (S.uiState !== \"partial\") {\n const hhiNow = hhiOf(rows);\n const mayorNow = rows[0] ? rows[0].pct : 0;\n const serie = D.DATES.map((_, i) => {\n if (i === D.DATES.length - 1) return S.concTR === \"hhi\" ? +hhiNow.toFixed(4) : +mayorNow.toFixed(2);\n const drift = 1 + Math.sin(i * 1.7) * 0.03 - (D.DATES.length - 1 - i) * 0.004;\n return S.concTR === \"hhi\" ? +(hhiNow * drift).toFixed(4) : +(mayorNow * drift).toFixed(2);\n });\n const lim = S.concTR === \"hhi\" ? D.LIMITES.hhiInterno : D.LIMITES.emisorPctInterno;\n const yMax = Math.max(lim * 1.08, Math.max(...serie) * 1.05);\n mkChart(\"chCTR\", {\n grid: { left: 56, right: 16, top: 24, bottom: 40 },\n tooltip: { trigger: \"axis\", valueFormatter: v => S.concTR === \"hhi\" ? v.toFixed(4) : fPct(v, 1) },\n xAxis: { type: \"category\", data: D.DATES, axisLabel: { fontSize: 9, rotate: 40 } },\n yAxis: { type: \"value\", scale: true, max: +yMax.toFixed(4), axisLabel: { formatter: v => S.concTR === \"hhi\" ? v.toFixed(3) : v + \"%\", fontSize: 10 } },\n series: [{ type: \"line\", symbol: \"circle\", symbolSize: 4, color: TK(\"--primary\"), data: serie,\n markLine: { silent: true, symbol: \"none\",\n lineStyle: { type: \"dashed\", color: TK(\"--sem-warn\") },\n label: { formatter: \"Límite interno\", fontSize: 9, color: TK(\"--sem-warn\") },\n data: [{ yAxis: lim }] } }],\n });\n }\n // BL — divergentes Δ por emisor (top-7 |Δ| + Otros + Total) — NEUTRO\n const tp = totalPrev();\n let items = rows.map(r => ({ k: r.emisor, d: r.vm - r.prev, dp: r.prev > 0 ? (r.vm - r.prev) / r.prev * 100 : null }));\n items.sort((a, b) => Math.abs(b.d) - Math.abs(a.d));\n const topB = items.slice(0, 7), restB = items.slice(7);\n items = [...topB];\n if (restB.length) items.push({ k: \"Otros\", d: restB.reduce((a, x) => a + x.d, 0), dp: null });\n items.push({ k: \"Total\", d: t - tp, dp: tp > 0 ? (t - tp) / tp * 100 : null });\n mkChart(\"chCBL\", {\n grid: { left: 60, right: 16, top: 28, bottom: 62 },\n tooltip: { trigger: \"axis\", valueFormatter: v => (v >= 0 ? \"+\" : \"−\") + fMM(Math.abs(v)).slice(1) },\n xAxis: { type: \"category\", data: items.map(i => i.k), axisLabel: { fontSize: 9, rotate: 38 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n series: [{ type: \"bar\", barWidth: 20,\n itemStyle: { color: p => items[p.dataIndex].k === \"Total\" ? TK(\"--secondary\") : items[p.dataIndex].k === \"Otros\" ? GRAY : TK(\"--primary\") },\n label: { show: true, position: \"top\", fontSize: 9, color: TK(\"--text-2\"),\n formatter: p => items[p.dataIndex].dp === null ? \"\" : (items[p.dataIndex].dp >= 0 ? \"+\" : \"\") + nf1.format(items[p.dataIndex].dp) + \"%\" },\n data: items.map(i => +i.d.toFixed(2)) }],\n });\n }\n\n function chartsMRFecha() {\n const a = mrAgg();\n // TL — barras por rating (orden de escala; cero real = 0; sin dato = ausente + N/D)\n const cats = D.RATINGS.filter(r => D.positions.some(p => p.rating === r));\n mkChart(\"chMF_TL\", {\n grid: { left: 60, right: 16, top: 24, bottom: 30 },\n tooltip: { trigger: \"axis\", valueFormatter: v => fMM(v) },\n xAxis: { type: \"category\", data: cats, axisLabel: { fontSize: 10 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n series: [{ type: \"bar\", barWidth: 26,\n itemStyle: { color: p => ratColor(cats[p.dataIndex]) },\n label: { show: true, position: \"top\", fontSize: 9, color: TK(\"--text-2\"), formatter: p => fMM(p.value) },\n data: cats.map(r => { const row = a.rows.find(x => x.r === r); return row ? +row.contrib.toFixed(2) : 0; }) }],\n });\n // TR — dona de contribución (Emisor top-7+Otros / Tipo) — prorrateo §2\n if (S.uiState !== \"partial\") {\n let entries = S.mrDona === \"Emisor\" ? mrEmisores() : mrTipos();\n if (entries.length > 8) {\n const top = entries.slice(0, 7), rest = entries.slice(7).reduce((x, e) => x + e[1], 0);\n entries = [...top, [\"Otros\", rest]];\n }\n mkChart(\"chMF_TR\", {\n tooltip: { trigger: \"item\", valueFormatter: v => fMM(v) },\n legend: { orient: \"vertical\", right: 0, top: \"middle\", textStyle: { fontSize: 10 },\n formatter: n => { const e = entries.find(x => x[0] === n); return `${n} ${fMM(e[1])}`; } },\n series: [{ type: \"pie\", radius: [\"52%\", \"74%\"], center: [\"32%\", \"50%\"],\n label: { formatter: p => fPct(p.percent, 1), fontSize: 10 },\n data: entries.map((e, i) => ({ name: e[0], value: +e[1].toFixed(2), itemStyle: { color: e[0] === \"Otros\" ? GRAY : CATEG[i % CATEG.length] } })) }],\n });\n }\n // BL — standalone vs contribución por rating (agrupadas; la brecha = diversificación)\n const cats2 = a.rows.map(x => x.r);\n mkChart(\"chMF_BL\", {\n grid: { left: 60, right: 16, top: 28, bottom: 46 },\n tooltip: { trigger: \"axis\", valueFormatter: v => fMM(v) },\n legend: { bottom: 0, itemWidth: 10, itemHeight: 10, textStyle: { fontSize: 10 } },\n xAxis: { type: \"category\", data: cats2, axisLabel: { fontSize: 10 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n series: [\n { name: S.medida + \" standalone\", type: \"bar\", barWidth: 14, color: TK(\"--secondary\"), data: a.rows.map(x => +x.stand.toFixed(2)) },\n { name: \"Contribución a cartera\", type: \"bar\", barWidth: 14, color: TK(\"--primary\"), data: a.rows.map(x => +x.contrib.toFixed(2)) },\n ],\n });\n }\n\n function chartsMRHist() {\n const series = mrHistSeries();\n const mk = mrMarkerIdx();\n const markLine = { silent: true, symbol: \"none\",\n lineStyle: { type: \"dashed\", color: TK(\"--text-2\"), width: 1.4 },\n label: { formatter: D.SUPUESTO_MARKERS[0].label, fontSize: 9, color: TK(\"--text-2\"), position: \"insideEndTop\" },\n data: [{ xAxis: D.DATES[mk] }] };\n // TL — área apilada por rating + marcador de cambio de supuesto\n mkChart(\"chMH_TL\", {\n grid: { left: 60, right: 16, top: 24, bottom: 44 },\n tooltip: { trigger: \"axis\", valueFormatter: v => fMM(v) },\n legend: { bottom: 0, itemWidth: 10, itemHeight: 10, textStyle: { fontSize: 10 } },\n xAxis: { type: \"category\", data: D.DATES, axisLabel: { fontSize: 9, rotate: 40 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n series: series.map((s, i) => ({\n name: s.r, type: \"line\", stack: \"mr\", areaStyle: { opacity: 0.85 }, symbol: \"none\",\n lineStyle: { width: 1 }, color: ratColor(s.r), data: s.data,\n ...(i === 0 ? { markLine } : {}),\n })),\n });\n // TR — razón /VM (%) + mismo marcador\n if (S.uiState !== \"partial\") {\n const totals = mrHistTotals();\n const vmSerie = [...D.HIST_TOTAL]; vmSerie[vmSerie.length - 1] = mrAgg().vmTot;\n const razon = D.DATES.map((_, i) => +(totals[i] / vmSerie[i] * 100).toFixed(3));\n mkChart(\"chMH_TR\", {\n grid: { left: 52, right: 16, top: 24, bottom: 40 },\n tooltip: { trigger: \"axis\", valueFormatter: v => fPct(v) },\n xAxis: { type: \"category\", data: D.DATES, axisLabel: { fontSize: 9, rotate: 40 } },\n yAxis: { type: \"value\", scale: true, axisLabel: { formatter: v => v + \"%\", fontSize: 10 } },\n series: [{ type: \"line\", symbol: \"circle\", symbolSize: 4, color: TK(\"--primary\"), data: razon, markLine }],\n });\n }\n // BL — divergentes Δ (Calificación / Emisor top-7 |Δ| + Otros) — NEUTRO\n const a = mrAgg(), ap = mrAgg(true);\n let items;\n if (S.mrBL === \"Calificación\") {\n items = a.rows.map(x => {\n const prev = ap.rows.find(y => y.r === x.r);\n const d = x.contrib - (prev ? prev.contrib : 0);\n return { k: x.r, d, dp: prev ? d / prev.contrib * 100 : null };\n });\n } else {\n const cur = new Map(mrEmisores()), prev = new Map(mrEmisores(true));\n items = [...cur.keys()].map(k => {\n const d = cur.get(k) - (prev.get(k) || 0);\n return { k, d, dp: prev.get(k) ? d / prev.get(k) * 100 : null };\n }).sort((x, y) => Math.abs(y.d) - Math.abs(x.d));\n const top = items.slice(0, 7), rest = items.slice(7);\n items = [...top];\n if (rest.length) items.push({ k: \"Otros\", d: rest.reduce((x, y) => x + y.d, 0), dp: null });\n }\n items.push({ k: \"Total\", d: a.total - ap.total, dp: ap.total > 0 ? (a.total - ap.total) / ap.total * 100 : null });\n mkChart(\"chMH_BL\", {\n grid: { left: 60, right: 16, top: 28, bottom: 56 },\n tooltip: { trigger: \"axis\", valueFormatter: v => (v >= 0 ? \"+\" : \"−\") + fMM(Math.abs(v)).slice(1) },\n xAxis: { type: \"category\", data: items.map(i => i.k), axisLabel: { fontSize: 9, rotate: 32 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => (v < 0 ? \"−\" : \"\") + \"$\" + (Math.abs(v) < 10 ? nf1.format(Math.abs(v)) : nf0(Math.abs(v))) + \"MM\", fontSize: 10 } },\n series: [{ type: \"bar\", barWidth: 22,\n itemStyle: { color: p => items[p.dataIndex].k === \"Total\" ? TK(\"--secondary\") : items[p.dataIndex].k === \"Otros\" ? GRAY : TK(\"--primary\") },\n label: { show: true, position: \"top\", fontSize: 9, color: TK(\"--text-2\"),\n formatter: p => items[p.dataIndex].dp === null ? \"\" : (items[p.dataIndex].dp >= 0 ? \"+\" : \"\") + nf1.format(items[p.dataIndex].dp) + \"%\" },\n data: items.map(i => +i.d.toFixed(2)) }],\n });\n }\n\n const nf0v = new Intl.NumberFormat(\"en-US\", { maximumFractionDigits: 0 });\n function nf0(v) { return nf0v.format(v); }\n\n // ---------- Backtesting › Resumen de Excepciones (§14) ----------\n const BT_TITLES = [\"Excepciones — VaR 99% · 1 año\", \"Kupiec POF\", \"Christoffersen\", \"Precisión MC de la cola\", \"Uso Límite\"];\n const BT_CONV = \"VaR 99%·1a escalado al horizonte entre cortes (raíz de tiempo, cortes EOD diarios) — convención del mock: la metodología oficial del tenant gobierna la comparación\";\n const BT_ROT_BASILEA = \"Veredicto estadístico de validación del modelo — no es cumplimiento de límites\";\n const ND_WIN = n => `N/D — ventana incompleta (${n}/250)`;\n function btWin() {\n const cs = D.BT.cortes;\n const val = cs.filter(c => c.valida);\n const excAll = cs.filter(c => c.pnl < -c.varEsc);\n return { cs, val, n: val.length, excAll, excVal: excAll.filter(c => c.valida) };\n }\n function btUsoLimite() {\n // VaR base FORZADO (§14): mismo total que Métricas de Riesgo bajo medida VaR — no sigue el selector global\n const lim = D.LIM_MEDIDA.VaR, total = D.BT.VAR1A;\n const uso = total / lim * 100;\n const st = uso > 100 ? [\"err\", \"Incumplimiento\"] : uso >= 90 ? [\"warn\", \"Advertencia\"] : [\"ok\", \"Ok\"];\n return kpiCard(\"Uso Límite\", `${fPct(uso, 1)} ${st[1]}`,\n `VaR base vs. límite interno ${fMM(lim)} · cartera consolidada completa · medida VaR fija — no sigue al selector de otras secciones`, []);\n }\n function btHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(BT_TITLES)}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(BT_TITLES)}

${statePanels(errBox())}

`;\n const w = btWin();\n const esp = (w.n * 0.01).toFixed(2);\n const excPrev = w.excVal.filter(c => c.f !== D.AS_OF).length; // ventana del corte anterior (mismo supuesto)\n const kpis =\n kpiCard(`Excepciones — VaR 99% · 1 año ⓘ`,\n `${w.excVal.length} en ${w.n} cortes`,\n `ventana válida, objetivo 250 · esperadas ~${esp} al 99% en ${w.n} · Consolidado en DOP · TC del corte`,\n [[\"anterior\", `${excPrev} (${w.n - 1}/250)`]]) +\n kpiCard(`Kupiec POF ⓘ`,\n \"N/D\", ND_WIN(w.n), [[\"anterior\", `N/D (${w.n - 1}/250)`]]) +\n kpiCard(`Christoffersen ⓘ`,\n \"N/D\", ND_WIN(w.n), [[\"anterior\", `N/D (${w.n - 1}/250)`]]) +\n kpiCard(`Precisión MC de la cola ⓘ`,\n `±${nf1.format(D.BT.mcPrecision)}%`,\n `del VaR @99% (cuantil oficial) · colas más profundas requieren ~1/q escenarios`, []) +\n btUsoLimite();\n return `

${kpis}

Serie de excepciones ⓘ

${BT_CONV}.

Series en neutro; la excepción se marca por forma (rombo) en un único color de alerta rotulado — nunca la terna semafórica. Cada punto consolidado al TC de su fecha.

${S.uiState === \"partial\" ? errBox() : btTRHtml(w)}

Convergencia del estimador MC ⓘ

\n ${[\"1%\", \"0.5%\", \"0.1%\"].map(q =>\n ``).join(\"\")}

Colas profundas: la banda de confianza escala ~1/q en escenarios.
Bandas del estimador Monte Carlo — no es la distribución de valor del portafolio.

Cuantil de diagnóstico: ${S.btQ} · eje Y fijo común entre cuantiles, a escala real — el ensanchamiento de la banda es el caveat.

Excepciones del VaR — 99% · 1 año (DOP consolidado)

${tablaBt(w)}

Historia visible completa — las excepciones fuera de la ventana válida van atenuadas y rotuladas (el KPI cuenta solo la ventana válida; la diferencia es legible, no un descuadre). ${BT_CONV}. Cada punto consolidado al TC de su fecha. Variaciones al TC de cada corte; incluyen efecto FX.

La unidad de análisis es el corte del portafolio completo — no hay vistas por emisor y la leyenda de prorrateo no aplica. VaR y P&L en DOP consolidado, jamás partidos por moneda; el filtro de moneda no aplica en esta sección.

`;\n }\n function btTRHtml(w) {\n const pv = D.BT.prevWin;\n return `

Validación estadística ⓘ

Ventana válida: ${w.n} cortes desde ${w.val[0].f} (objetivo 250)

Kupiec POF

${ND_WIN(w.n)}

LR: N/D · χ²(1) · p-value: N/D · veredicto: N/D

Christoffersen (conjunto: independencia + cobertura)

${ND_WIN(w.n)}

LR conjunto: N/D · χ²(2) · p-value: N/D · veredicto: N/D

Semáforo de Basilea

${ND_WIN(w.n)}${BT_ROT_BASILEA}

Zonas normalizadas a 250 cortes: verde 0–4 · amarilla 5–9 · roja 10+ · multiplicador base 3 + recargo (referencia ilustrativa del marco de mercado)

Ventana anterior (cerrada) — referencia histórica, no es el veredicto vigente

${pv.rot} · ${pv.exc} excepciones

\n Kupiec POF${pv.kupiec.v}LR ${pv.kupiec.lr} · χ²(1) · p=${pv.kupiec.p}\n Christoffersen (conjunto)${pv.christo.v}LR ${pv.christo.lr} · χ²(2) · p=${pv.christo.p}\n

Zona ${pv.basilea.zona} · ${pv.basilea.conteo}${BT_ROT_BASILEA}

Multiplicador ${pv.basilea.mult} (referencia ilustrativa del marco de mercado)

Excepciones contadas bajo la convención de escala del mock. Las pruebas se computan solo sobre la sub-ventana homogénea vigente — jamás cruzan un cambio de supuesto (§2).

`;\n }\n function tablaBt(w) {\n const rows = w.excAll.map(c => {\n const exceso = -(c.pnl + c.varEsc);\n const sup = c.valida ? \"S&P · ρ=25%\" : \"S&P · ρ=20%\";\n return `\n ${c.f}${c.valida ? \"\" : ' fuera de ventana'}\n −${nf2.format(Math.abs(c.pnl))}MM\n ${nf2.format(c.varEsc)}MM\n ${nf2.format(exceso)}MM\n ${sup}${c.valida ? \"Sí\" : \"No\"}`;\n }).join(\"\");\n return `\n \n ${rows}

Fecha	P&L realizado (DOP MM)	VaR 99%·1a escalado (DOP MM)	Exceso (DOP MM)	Supuesto vigente	Ventana válida

`;\n }\n function chartsBt() {\n const w = btWin();\n const dates = w.cs.map(c => c.f);\n const excPts = w.cs.map((c, i) => [c, i]).filter(([c]) => c.pnl < -c.varEsc);\n mkChart(\"chBT_TL\", {\n grid: { left: 56, right: 16, top: 30, bottom: 64 },\n tooltip: { trigger: \"axis\", valueFormatter: v => (v < 0 ? \"−\" : \"\") + \"$\" + nf1.format(Math.abs(v)) + \"MM\" },\n legend: { top: 0, itemWidth: 10, itemHeight: 10, textStyle: { fontSize: 10 }, data: [\"P&L realizado\", \"−VaR escalado\", \"Excepción\"] },\n dataZoom: [{ type: \"slider\", height: 14, bottom: 4 }],\n xAxis: { type: \"category\", data: dates, axisLabel: { fontSize: 9, rotate: 40 } },\n yAxis: { type: \"value\", axisLabel: { formatter: v => (v < 0 ? \"−\" : \"\") + \"$\" + nf0(Math.abs(v)) + \"MM\", fontSize: 10 } },\n series: [\n { name: \"P&L realizado\", type: \"line\", symbol: \"none\", lineStyle: { width: 1.1 }, color: TK(\"--primary\"), data: w.cs.map(c => c.pnl) },\n { name: \"−VaR escalado\", type: \"line\", symbol: \"none\", lineStyle: { width: 1.4, type: \"dashed\" }, color: TK(\"--secondary\"), data: w.cs.map(c => -c.varEsc),\n markLine: { silent: true, symbol: \"none\", lineStyle: { type: \"dashed\", color: TK(\"--text-2\"), width: 1.4 },\n label: { formatter: D.BT.cambioLabel, fontSize: 9, color: TK(\"--text-2\"), position: \"insideEndTop\" },\n data: [{ xAxis: D.BT.EFFECTIVE }] } },\n { name: \"Excepción\", type: \"scatter\", symbol: \"diamond\", symbolSize: 11, color: TK(\"--sem-warn\"), data: excPts.map(([c, i]) => [i, c.pnl]) },\n ],\n });\n // BL — convergencia MC: eje Y fijo COMÚN entre cuantiles (min/max globales, sin recorte)\n const Ns = [5, 10, 15, 20, 30, 40, 50, 65, 80, 100]; // miles de escenarios\n const base = D.BT.VAR1A, K = 1.017;\n let yMin = Infinity, yMax = -Infinity;\n Object.values(D.BT.quantiles).forEach(({ q, nivel }) => {\n const v = base * nivel;\n const h0 = K * v * Math.sqrt((1 - q) / (q * Ns[0] * 1000));\n yMin = Math.min(yMin, v - h0 - 6); yMax = Math.max(yMax, v + h0 + 6);\n });\n const sel = D.BT.quantiles[S.btQ], nivel = base * sel.nivel;\n const est = Ns.map((n, i) => +(nivel * (1 + 0.02 * Math.sin(i * 2.1) * Math.exp(-i / 3.2))).toFixed(2));\n const hw = Ns.map(n => K * nivel * Math.sqrt((1 - sel.q) / (sel.q * n * 1000)));\n const lower = est.map((e, i) => +(e - hw[i]).toFixed(2));\n const band = hw.map(h => +(2 * h).toFixed(2));\n mkChart(\"chBT_BL\", {\n grid: { left: 56, right: 16, top: 30, bottom: 40 },\n tooltip: { trigger: \"axis\", formatter: ps => {\n const i = ps[0].dataIndex;\n return `${Ns[i]}k escenarios
VaR estimado: ${fMM(est[i])}
Banda: ±${fMM(hw[i]).slice(1)} (±${nf1.format(hw[i] / est[i] * 100)}%)`;\n } },\n legend: { top: 0, itemWidth: 10, itemHeight: 10, textStyle: { fontSize: 10 }, data: [`VaR estimado @${S.btQ}`, \"Banda de confianza\"] },\n xAxis: { type: \"category\", data: Ns.map(n => n + \"k\"), axisLabel: { fontSize: 10 }, name: \"escenarios\", nameLocation: \"middle\", nameGap: 26, nameTextStyle: { fontSize: 9 } },\n yAxis: { type: \"value\", min: Math.floor(yMin), max: Math.ceil(yMax), axisLabel: { formatter: v => \"$\" + nf0(v) + \"MM\", fontSize: 10 } },\n series: [\n { name: \"límite inferior\", type: \"line\", stack: \"band\", symbol: \"none\", lineStyle: { opacity: 0 }, color: \"transparent\", data: lower, silent: true },\n { name: \"Banda de confianza\", type: \"line\", stack: \"band\", symbol: \"none\", lineStyle: { opacity: 0 }, areaStyle: { color: TK(\"--primary\"), opacity: 0.16 }, color: TK(\"--primary\"), data: band },\n { name: `VaR estimado @${S.btQ}`, type: \"line\", symbol: \"circle\", symbolSize: 5, color: TK(\"--secondary\"), lineStyle: { width: 1.6 }, data: est },\n ],\n });\n }\n\n // ---------- Export CSV (pies + supuesto serializados — X-8d) ----------\n function exportCSV(which) {\n const lines = [];\n if (which === \"backtesting\") {\n const w = btWin();\n lines.push(`\"Risk Intelligence — Crédito · Backtesting (Excepciones del VaR — 99% · 1 año, DOP consolidado)\";\"as_of ${D.AS_OF}\";\"Supuesto: ${D.SUPUESTO.fuente} · ${D.SUPUESTO.metodo} · ${D.SUPUESTO.tc}\";\"Ventana válida: ${w.n} cortes desde ${w.val[0].f} (objetivo 250)\"`);\n lines.push(`\"Fecha\";\"P&L realizado (DOP MM)\";\"VaR 99%·1a escalado (DOP MM)\";\"Exceso (DOP MM)\";\"Supuesto vigente\";\"Ventana válida\"`);\n w.excAll.forEach(c => {\n lines.push(`\"${c.f}\";${c.pnl.toFixed(2)};${c.varEsc.toFixed(2)};${(-(c.pnl + c.varEsc)).toFixed(2)};\"${c.valida ? \"S&P · ρ=25%\" : \"S&P · ρ=20%\"}\";\"${c.valida ? \"Sí\" : \"No\"}\"`);\n });\n lines.push(`\"Nota\";\"${BT_CONV}.\"`);\n lines.push(`\"Nota\";\"El KPI de excepciones cuenta solo la ventana válida; esta tabla lista la historia visible completa — las filas fuera de ventana van rotuladas.\"`);\n lines.push(`\"Nota\";\"Cada punto consolidado al TC de su fecha. Variaciones al TC de cada corte; incluyen efecto FX. VaR y P&L jamás se parten por moneda.\"`);\n lines.push(`\"Nota\";\"Colas profundas: la banda de confianza escala ~1/q en escenarios. Bandas del estimador Monte Carlo — no es la distribución de valor del portafolio.\"`);\n const blobB = new Blob([\"\\uFEFF\" + lines.join(\"\\n\")], { type: \"text/csv;charset=utf-8\" });\n const aB = document.createElement(\"a\");\n aB.href = URL.createObjectURL(blobB); aB.download = \"credito-backtesting-excepciones.csv\"; aB.click();\n return;\n }\n if (which === \"factores\") {\n const a = frAgg(), ap = frAgg(true), comp = D.FR.prevComparable;\n const dc = (cur, prev) => comp && prev > 0 ? [(cur - prev).toFixed(2), ((cur - prev) / prev * 100).toFixed(2)] : ['\"N/D\"', '\"N/D\"'];\n lines.push(`\"Risk Intelligence — Crédito · Factores de Riesgo (Contribución al VaR — 99% · 1 año, DOP consolidado)\";\"as_of ${D.AS_OF}\";\"Supuesto: ${D.SUPUESTO.fuente} · ${D.SUPUESTO.metodo} · ${D.SUPUESTO.tc}\"`);\n lines.push(`\"Factor\";\"Calificación\";\"Contribución (DOP MM)\";\"% del VaR\";\"Var (DOP MM)\";\"Var %\";\"Convención\"`);\n FR_FACTORES.forEach(([f, lbl]) => {\n const d1 = dc(a.fact[f], ap.fact[f]);\n lines.push(`\"${lbl}\";\"—\";${a.fact[f].toFixed(2)};${(a.fact[f] / a.total * 100).toFixed(2)};${d1[0]};${d1[1]};\"Euler (aditiva)\"`);\n a.rows.forEach(x => {\n const p = ap.rows.find(y => y.r === x.r);\n const d2 = dc(x.cells[f], p ? p.cells[f] : 0);\n lines.push(`\"${lbl}\";\"${x.r}\";${x.cells[f].toFixed(2)};${(x.cells[f] / a.total * 100).toFixed(2)};${d2[0]};${d2[1]};\"Euler (aditiva)\"`);\n });\n });\n const dcor = dc(a.corr, ap.corr);\n lines.push(`\"Efecto correlación\";\"—\";${a.corr.toFixed(2)};${(a.corr / a.total * 100).toFixed(2)};${dcor[0]};${dcor[1]};\"Incremental (no aditiva — fuera del subtotal Euler)\"`);\n const dtot = dc(a.total, ap.total);\n lines.push(`\"Total (Σ Euler)\";\"—\";${a.total.toFixed(2)};100.00;${dtot[0]};${dtot[1]};\"Euler (aditiva)\"`);\n lines.push(`\"Nota\";\"${FR_CONV}.\"`);\n lines.push(`\"Nota\";\"${FR_GAUSS}.\"`);\n lines.push(`\"Nota\";\"Variaciones al TC de cada corte; incluyen efecto FX. La contribución va en DOP consolidado y jamás se parte por moneda. Totales calculados sobre valores sin redondear. Δ solo entre cortes con el mismo supuesto.\"`);\n const blobF = new Blob([\"\\uFEFF\" + lines.join(\"\\n\")], { type: \"text/csv;charset=utf-8\" });\n const aF = document.createElement(\"a\");\n aF.href = URL.createObjectURL(blobF); aF.download = \"credito-factores-mapa.csv\"; aF.click();\n return;\n }\n if (which === \"marginal\") {\n lines.push(`\"Risk Intelligence — Crédito · Marginal (${S.medida})\";\"as_of ${D.AS_OF}\";\"Supuesto: ${D.SUPUESTO.fuente} · ${D.SUPUESTO.metodo} · ${D.SUPUESTO.tc}\"`);\n lines.push(`\"Calificación\";\"Emisor\";\"Instrumento\";\"Moneda\";\"VM DOP (MM)\";\"${S.medida} Marginal (DOP MM)\";\"Componental (DOP MM)\";\"% del total (base componental)\";\"Standalone (DOP MM)\";\"Benef. diversificación (DOP MM)\";\"Var Marginal (DOP MM)\";\"Var %\"`);\n const rows = mgRows(), prevRows = mgRows(true);\n const total = rows.reduce((a, x) => a + x.comp, 0);\n D.RATINGS.forEach(r => rows.filter(x => x.p.rating === r).sort((a, b) => b.v - a.v).forEach(x => {\n const prev = prevRows.find(y => y.p.inst === x.p.inst && y.p.emisor === x.p.emisor);\n const dv = prev ? (x.marg - prev.marg).toFixed(2) : '\"N/D\"';\n const dvp = prev && prev.marg !== 0 ? ((x.marg - prev.marg) / Math.abs(prev.marg) * 100).toFixed(2) : '\"N/D\"';\n lines.push(`\"${x.p.rating}\";\"${x.p.emisor}\";\"${x.p.inst}\";\"${x.p.moneda}\";${x.v.toFixed(2)};${x.marg.toFixed(2)};${x.comp.toFixed(2)};${(x.comp / total * 100).toFixed(2)};${x.stand.toFixed(2)};${(x.stand - x.comp).toFixed(2)};${dv};${dvp}`);\n }));\n lines.push(`\"Nota\";\"${NOTA_GRANULARIDAD}\"`);\n lines.push(`\"Nota\";\"Marginal = leave-one-out, NO aditiva (Σ marginales ≠ ${S.medida} total). Componental = Euler, aditiva (Σ = ${S.medida} total). Los % pueden superar 100% por contribuciones negativas.\"`);\n lines.push(`\"Nota\";\"Var refiere al Marginal; variaciones al TC de cada corte, incluyen efecto FX. El ${S.medida} va en DOP consolidado y jamás se parte por moneda.\"`);\n const blobM = new Blob([\"\\uFEFF\" + lines.join(\"\\n\")], { type: \"text/csv;charset=utf-8\" });\n const aM = document.createElement(\"a\");\n aM.href = URL.createObjectURL(blobM); aM.download = `credito-marginal-${S.medida.toLowerCase()}.csv`; aM.click();\n return;\n }\n if (which === \"mr-fecha\" || which === \"mr-hist\") {\n // X-13: el export histórico serializa el supuesto de AMBOS cortes\n const supHist = which === \"mr-hist\" ? `;\"Supuesto corte anterior (28/01/2026): idéntico — ${D.SUPUESTO.fuente} · ${D.SUPUESTO.metodo}\"` : \"\";\n lines.push(`\"Risk Intelligence — Crédito · Métricas de Riesgo (${S.medida})\";\"as_of ${D.AS_OF}\";\"Supuesto: ${D.SUPUESTO.fuente} · ${D.SUPUESTO.metodo} · ${D.SUPUESTO.tc}\"${supHist}`);\n const a = mrAgg(), ap = mrAgg(true);\n if (which === \"mr-fecha\") {\n lines.push(`\"Calificación\";\"VM DOP (MM)\";\"VM USD (MM)\";\"VM Total DOP (MM)\";\"${S.medida} (DOP MM)\";\"% del ${S.medida}\"`);\n a.rows.forEach(x => {\n const ps = D.positions.filter(p => p.rating === x.r);\n const vmDop = ps.filter(p => p.moneda === \"DOP\").reduce((s, p) => s + p.vm, 0);\n const vmUsd = ps.filter(p => p.moneda === \"USD\").reduce((s, p) => s + p.vm, 0);\n lines.push(`\"${x.r}\";${vmDop ? vmDop.toFixed(2) : '\"—\"'};${vmUsd ? vmUsd.toFixed(2) : '\"—\"'};${x.vm.toFixed(2)};${x.contrib.toFixed(2)};${(x.contrib / a.total * 100).toFixed(2)}`);\n });\n } else {\n lines.push(`\"Calificación\";\"VM Total DOP (MM)\";\"${S.medida} (DOP MM)\";\"Var (DOP MM)\";\"Var %\"`);\n a.rows.forEach(x => {\n const prev = ap.rows.find(y => y.r === x.r);\n const dv = prev ? (x.contrib - prev.contrib).toFixed(2) : '\"N/D\"';\n const dvp = prev ? ((x.contrib - prev.contrib) / prev.contrib * 100).toFixed(2) : '\"N/D\"';\n lines.push(`\"${x.r}\";${x.vm.toFixed(2)};${x.contrib.toFixed(2)};${dv};${dvp}`);\n });\n lines.push(`\"Nota\";\"Variaciones al TC de cada corte; incluyen efecto FX. Δ solo entre cortes con el mismo supuesto.\"`);\n }\n lines.push(`\"Nota\";\"El ${S.medida} va en DOP consolidado y jamás se parte por moneda (no aditivo). Contribuciones por emisor: prorrateo por exposición — el modelo opera a nivel rating.\"`);\n const blob0 = new Blob([\"\\uFEFF\" + lines.join(\"\\n\")], { type: \"text/csv;charset=utf-8\" });\n const a0 = document.createElement(\"a\");\n a0.href = URL.createObjectURL(blob0); a0.download = `credito-metricas-${S.medida.toLowerCase()}-${which}.csv`; a0.click();\n return;\n }\n lines.push(`\"Risk Intelligence — Crédito · Portafolio\";\"as_of ${D.AS_OF}\";\"Supuesto: ${D.SUPUESTO.fuente} · ${D.SUPUESTO.metodo} · ${D.SUPUESTO.tc}\"`);\n if (which === \"global\") {\n lines.push(`\"Calificación\";\"VM DOP (MM)\";\"VM USD (MM)\";\"VM Total DOP (MM)\";\"% Global\";\"Var (DOP MM)\";\"Var %\"`);\n const t = total();\n D.RATINGS.forEach(r => {\n const ps = filt().filter(p => p.rating === r); if (!ps.length) return;\n const dopV = ps.filter(p => p.moneda === \"DOP\").reduce((a, p) => a + p.vm, 0);\n const usdV = ps.filter(p => p.moneda === \"USD\").reduce((a, p) => a + p.vm, 0);\n const tot = ps.reduce((a, p) => a + dop(p), 0), prev = ps.reduce((a, p) => a + dopPrev(p), 0);\n lines.push(`\"${r}\";${dopV ? dopV.toFixed(2) : '\"—\"'};${usdV ? usdV.toFixed(2) : '\"—\"'};${tot.toFixed(2)};${(tot / t * 100).toFixed(2)};${prev > 0 ? (tot - prev).toFixed(2) : '\"N/D\"'};${prev > 0 ? ((tot - prev) / prev * 100).toFixed(2) : '\"N/D\"'}`);\n });\n lines.push(`\"Nota\";\"Variaciones al TC de cada corte; incluyen efecto FX.\"`);\n } else {\n lines.push(`\"Emisor\";\"Moneda\";\"VM DOP (MM)\";\"% VM\";\"Aporte HHI\";\"Límite interno %\";\"Holgura/Exceso pp\";\"Estado\"`);\n emisorStats().forEach(r => {\n const st = estadoEmisor(r.pct);\n lines.push(`\"${r.emisor}\";\"${r.chip}\";${r.vm.toFixed(2)};${r.pct.toFixed(2)};${r.hhi.toFixed(4)};${D.LIMITES.emisorPctInterno.toFixed(1)};${(D.LIMITES.emisorPctInterno - r.pct).toFixed(1)};\"${st.txt}\"`);\n });\n lines.push(`\"Nota\";\"Límites de exposición sobre VM observado; sin límite → N/D.\"`);\n }\n const blob = new Blob([\"\\uFEFF\" + lines.join(\"\\n\")], { type: \"text/csv;charset=utf-8\" });\n const a = document.createElement(\"a\");\n a.href = URL.createObjectURL(blob); a.download = `credito-portafolio-${which}.csv`; a.click();\n }\n\n // ---------- Bind + estados ----------\n // ---------- Factores de Riesgo (F2.8, §12) — medida canónica VaR 99%·1a FORZADA (X-#1/#3): jamás S.medida ----------\n const FR_FACTORES = [[\"mig\", \"Migración (non-default)\"], [\"def\", \"Default\"], [\"rec\", \"Recovery\"], [\"spr\", \"Spread\"]];\n const FR_CONV = \"Convención de atribución: las barras/filas Euler (migración · default · recovery · spread) son ADITIVAS — Σ = VaR total. El Efecto correlación es INCREMENTAL (VaR(ρ activo) − VaR(ρ=0), método de diferencias) y jamás se suma con las barras: la correlación es la estructura que acopla los factores, no un sumando\";\n const FR_GAUSS = \"Retorno de activo estandarizado bajo Alt 2 — no es la distribución de valor del portafolio\";\n function frAgg(prev = false) {\n const getVal = prev ? dopPrev : dop;\n const vm = new Map();\n D.positions.forEach(p => vm.set(p.rating, (vm.get(p.rating) || 0) + getVal(p)));\n const rows = D.RATINGS.filter(r => (vm.get(r) || 0) > 0).map(r => {\n const contrib = vm.get(r) * D.RISK.VaR.rate[r] * D.RISK.VaR.divF[r]; // fórmula única sin redondear (≡ mrAgg-VaR ≡ BT.VAR1A)\n const cells = {}; FR_FACTORES.forEach(([f]) => { cells[f] = contrib * D.FR.W[r][f]; });\n return { r, contrib, cells };\n });\n const total = rows.reduce((s, x) => s + x.contrib, 0);\n const fact = {}; FR_FACTORES.forEach(([f]) => { fact[f] = rows.reduce((s, x) => s + x.cells[f], 0); });\n return { rows, total, fact, corr: total * (1 - D.FR.RHO_ZERO_FACTOR) };\n }\n function frPredominante() { // misma definición que §7.1 KPI 4: share de VM, cartera completa (X-#4)\n const m = new Map();\n D.positions.forEach(p => m.set(p.rating, (m.get(p.rating) || 0) + dop(p)));\n const top = [...m.entries()].sort((a, b) => b[1] - a[1])[0][0];\n return [\"CCC\", \"D\"].includes(top) ? \"CCC\" : top; // grupo predominante CCC–D → CCC (§12)\n }\n const frRatingSel = () => S.frRating || frPredominante();\n const FR_TITLES = [\"VaR de Crédito Total — 99% · 1 año\", \"Factor dominante\", \"Efecto correlación\", \"Probabilidad de ≥1 default\", \"Uso Límite\"];\n const kpisNDFR = () => FR_TITLES.map(t => kpiCard(t, \"N/D\", \"\", [])).join(\"\");\n function frUsoLimite(total) { // límite del VaR, nunca S.medida (X-#1; precedente btUsoLimite)\n const lim = D.LIM_MEDIDA.VaR;\n if (lim == null) return kpiCard(\"Uso Límite\", \"N/D\", \"Sin límite configurado para VaR\", []);\n const uso = total / lim * 100;\n const st = uso > 100 ? [\"err\", \"Incumplimiento\"] : uso >= 90 ? [\"warn\", \"Advertencia\"] : [\"ok\", \"Ok\"];\n return kpiCard(\"Uso Límite\", `${fPct(uso, 1)} ${st[1]}`,\n `VaR 99%·1a vs. límite interno ${fMM(lim)} · cartera consolidada completa`, []);\n }\n function frLeyendaTL(a) { // conmutación §12: jamás se afirma Σ=VaR con datos incompletos (rama no demostrable en el mock — feedback-log)\n const resid = a.total - FR_FACTORES.reduce((s, [f]) => s + (Number.isFinite(a.fact[f]) ? a.fact[f] : 0), 0);\n return Math.abs(resid) < 0.005\n ? \"Leyenda: Atribución Euler — Σ barras = VaR total.\"\n : `Leyenda: Atribución parcial — residual no atribuido: ${fMM(resid)}. Barra sin dato: ausente + “N/D”.`;\n }\n function frHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDFR()}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDFR()}

${statePanels(errBox())}

`;\n const a = frAgg(), ap = frAgg(true), comp = D.FR.prevComparable;\n const dv = a.total - ap.total, dvp = ap.total > 0 ? dv / ap.total * 100 : null;\n const orden = FR_FACTORES.slice();\n const dom = orden.sort((x, y) => a.fact[y[0]] - a.fact[x[0]])[0];\n const domPrev = FR_FACTORES.slice().sort((x, y) => ap.fact[y[0]] - ap.fact[x[0]])[0];\n const tipComp = \"Comparabilidad de cortes (§11): si el supuesto del corte anterior difiere, el valor anterior pasa a N/D — un factor dominante calculado bajo otro ρ no es comparable.\";\n const kpis =\n kpiCard(\"VaR de Crédito Total — 99% · 1 año\", fMM(a.total), \"Consolidado en DOP · TC del corte\", comp ? [\n [\"anterior\", fMM(ap.total)],\n [\"Δ\", (dv >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dv)).slice(1), \"trend\"],\n [\"Δ%\", dvp === null ? \"N/D\" : fPct(dvp), \"trend\"]] : [[\"anterior\", \"N/D\"], [\"Δ\", \"N/D\", \"trend\"], [\"Δ%\", \"N/D\", \"trend\"]]) +\n kpiCard(\"Factor dominante\", esc(dom[1].replace(\" (non-default)\", \"\")),\n `${fPct(a.fact[dom[0]] / a.total * 100, 1)} del VaR · atribución Euler`, [\n [\"anterior\", comp ? `${esc(domPrev[1].replace(\" (non-default)\", \"\"))}` : `N/D`]]) +\n kpiCard(`Efecto correlación ⓘ`,\n \"+\" + fMM(a.corr), `${fPct(a.corr / a.total * 100, 1)} del VaR · vs. ρ=0`, comp ? [\n [\"anterior\", \"+\" + fMM(ap.corr)],\n [\"Δ\", (a.corr - ap.corr >= 0 ? \"+\" : \"−\") + fMM(Math.abs(a.corr - ap.corr)).slice(1), \"trend\"]] : [[\"anterior\", \"N/D\"], [\"Δ\", \"N/D\", \"trend\"]]) +\n (comp\n ? kpiCard(`Probabilidad de ≥1 default ⓘ`,\n fPct(D.FR.P_DEFAULT * 100, 1), \"horizonte 1 año · libro completo\", [])\n : kpiCard(`Probabilidad de ≥1 default ⓘ`, \"N/D\", \"supuesto del corte anterior distinto\", [])) +\n frUsoLimite(a.total);\n const r0 = frRatingSel();\n return `

${kpis}

Contribución por factor — Atribución Euler (VaR, DOP MM)

${frLeyendaTL(a)} El “Efecto correlación” es el marcador anotado — nunca una barra. ${FR_CONV}.

${S.uiState === \"partial\" ? errBox() : `\n

Umbrales de Merton — variable latente

\n ${[\"AAA\", \"AA\", \"A\", \"BBB\", \"BB\", \"B\", \"CCC\"].map(r =>\n ``).join(\"\")}

${FR_GAUSS} — la silueta multimodal del valor vive en Escenarios.

Densidad normal estándar del retorno de activo con las fronteras Z del rating de partida ${r0} (default del selector: calificación predominante por share de VM). Bandas de llegada AAA…CCC y cola de default sombreada (inferior). D excluido del selector: estado absorbente, sin migración de salida. Umbrales derivados de la matriz de transición del supuesto.

Matriz ρ asumida (correlación de activos)

Matriz uniforme — ρ=25% axiomático (Alt 2). Escala secuencial neutra: la correlación no es estado de límite. Función del panel: auditar el supuesto — uniforme ⇒ método axiomático vigente; bajo cópula empírica (Alt 1) mostraría la co-migración por par de ratings y esta anotación conmutaría.

Contribución al VaR — 99% · 1 año (DOP consolidado)

${tablaFr(a, ap)}

Variaciones al TC de cada corte; incluyen efecto FX — la contribución va en DOP consolidado y jamás se parte por moneda. Totales calculados sobre valores sin redondear. ${FR_CONV}.

`;\n }\n function tablaFr(a, ap) {\n const comp = D.FR.prevComparable;\n const varCells = (cur, prev) => {\n if (!comp || !(prev > 0)) return `N/DN/D`;\n const d = cur - prev;\n return `${(d >= 0 ? \"+\" : \"−\") + fMM(Math.abs(d)).slice(1)}${fPct(d / prev * 100)}`;\n };\n let rows = \"\";\n FR_FACTORES.forEach(([f, lbl]) => {\n const open = S.frExp[f];\n rows += `${open ? \"▾\" : \"▸\"} ${lbl}\n ${fMM(a.fact[f])}${fPct(a.fact[f] / a.total * 100, 1)}${varCells(a.fact[f], ap.fact[f])}`;\n if (open) a.rows.forEach(x => {\n const p = ap.rows.find(y => y.r === x.r);\n rows += `${x.r}\n ${fMM(x.cells[f])}${fPct(x.cells[f] / a.total * 100, 1)}${varCells(x.cells[f], p ? p.cells[f] : 0)}`;\n });\n });\n rows += `Efecto correlación incremental\n +${fMM(a.corr)}${fPct(a.corr / a.total * 100, 1)}${varCells(a.corr, ap.corr)}`;\n rows += `Total (Σ Euler)${fMM(a.total)}100.0%${varCells(a.total, ap.total)}`;\n return `${rows}

Factor / Calificación	Contribución (DOP MM)	% del VaR	Var (DOP MM)	Var %

`;\n }\n function chartsFr() {\n const a = frAgg();\n mkChart(\"chFR_TL\", {\n grid: { left: 56, right: 16, top: 40, bottom: 30 },\n tooltip: { trigger: \"axis\", valueFormatter: v => fMM(v) },\n xAxis: { type: \"category\", data: FR_FACTORES.map(x => x[1].replace(\" (non-default)\", \"\")), axisLabel: { fontSize: 10, interval: 0 } },\n yAxis: { type: \"value\", name: \"DOP MM\", axisLabel: { formatter: v => \"$\" + nf0(v) } },\n series: [{\n type: \"bar\", name: \"Contribución Euler\", data: FR_FACTORES.map(([f]) => +a.fact[f].toFixed(2)), barWidth: \"46%\",\n itemStyle: { color: TK(\"--primary\") },\n label: { show: true, position: \"top\", fontSize: 10, formatter: p => fMM(p.value) },\n markLine: {\n symbol: \"none\", lineStyle: { type: \"dashed\", color: TK(\"--text-2\"), width: 1.5 },\n label: { formatter: `Efecto correlación: +${fMM(a.corr)} (incremental, no aditivo)`, fontSize: 10, position: \"insideEndTop\", color: TK(\"--text-2\") },\n data: [{ yAxis: +a.corr.toFixed(2) }],\n },\n }],\n });\n if (S.uiState !== \"partial\") {\n const bnds = D.FR.Z[frRatingSel()]; // peor→mejor: D, CCC, …, AA (zUp crecientes)\n const xs = []; for (let z = -4; z <= 4.0001; z += 0.05) xs.push(+z.toFixed(2));\n const areas = []; let lo = -4;\n bnds.forEach(b => { if (b.zUp > -4) { areas.push({ name: b.dest, lo, hi: Math.min(b.zUp, 4) }); lo = b.zUp; } });\n areas.push({ name: \"AAA\", lo, hi: 4 });\n mkChart(\"chFR_TR\", {\n grid: { left: 30, right: 16, top: 24, bottom: 30 },\n tooltip: { show: false },\n xAxis: { type: \"value\", min: -4, max: 4, name: \"Z\", nameGap: 6, axisLabel: { fontSize: 10 } },\n yAxis: { type: \"value\", axisLabel: { show: false }, splitLine: { show: false } },\n series: [{\n type: \"line\", data: xs.map(z => [z, +(Math.exp(-z * z / 2) / Math.sqrt(2 * Math.PI)).toFixed(5)]),\n showSymbol: false, lineStyle: { color: TK(\"--primary\"), width: 2 },\n areaStyle: { color: TK(\"--primary\"), opacity: 0.05 },\n markArea: {\n silent: true,\n data: areas.map((ar, i) => [\n { xAxis: ar.lo, itemStyle: { color: ar.name === \"D\" ? GRAY : \"transparent\", opacity: ar.name === \"D\" ? 0.45 : 0 },\n label: { formatter: ar.name === \"D\" ? \"D (default)\" : ar.name, fontSize: 8, rotate: 90,\n position: (ar.hi - ar.lo) < 0.45 ? (i % 2 ? \"insideTop\" : \"insideBottom\") : \"insideBottom\",\n color: TK(\"--text-2\") } },\n { xAxis: ar.hi }]),\n },\n markLine: {\n symbol: \"none\", lineStyle: { color: TK(\"--text-2\"), width: 1, opacity: 0.5 },\n label: { show: false },\n data: bnds.filter(b => b.zUp > -4 && b.zUp < 4).map(b => ({ xAxis: b.zUp })),\n },\n }],\n });\n }\n const L7 = [\"AAA\", \"AA\", \"A\", \"BBB\", \"BB\", \"B\", \"CCC–D\"];\n const hm = []; for (let i = 0; i < 7; i++) for (let j = 0; j < 7; j++) hm.push([j, i, D.FR.RHO_UNIF]);\n mkChart(\"chFR_BL\", {\n grid: { left: 56, right: 16, top: 10, bottom: 30 },\n tooltip: { formatter: p => `${L7[p.value[1]]} × ${L7[p.value[0]]}: ρ = ${p.value[2]}%
Correlación entre pares de emisores de los ratings — no es autocorrelación de un activo (≠1 por diseño)` },\n xAxis: { type: \"category\", data: L7, axisLabel: { fontSize: 10 } },\n yAxis: { type: \"category\", data: L7, axisLabel: { fontSize: 10 } },\n visualMap: { show: false, min: 0, max: 50, inRange: { color: [\"#eef3fa\", \"#a9c7ea\", TK(\"--primary\")] } },\n series: [{ type: \"heatmap\", data: hm, label: { show: true, fontSize: 9, formatter: p => p.value[2] }, itemStyle: { borderColor: \"#fff\", borderWidth: 1 } }],\n });\n }\n\n function bindBody(section, tab) {\n document.querySelectorAll(\"[data-retry]\").forEach(b => b.onclick = () => { S.uiState = \"ok\"; render(); });\n document.querySelectorAll(\"[data-export]\").forEach(b => b.onclick = () => exportCSV(b.dataset.export));\n if (S.uiState === \"loading\" || S.uiState === \"empty\" || S.uiState === \"error\") return;\n if (section.id === \"metricas-de-riesgo\") {\n if (tab.id === \"a-la-fecha\") {\n const sd = $(\"#selMrDona\"); if (sd) sd.onchange = e => { S.mrDona = e.target.value; render(); };\n document.querySelectorAll(\"tr.grp[data-mrf]\").forEach(r => r.onclick = () => { S.mrExp[r.dataset.mrf] = !S.mrExp[r.dataset.mrf]; render(); });\n chartsMRFecha();\n } else {\n const sb = $(\"#selMrBL\"); if (sb) sb.onchange = e => { S.mrBL = e.target.value; render(); };\n document.querySelectorAll(\"tr.grp[data-mrh]\").forEach(r => r.onclick = () => { S.mrExpH[r.dataset.mrh] = !S.mrExpH[r.dataset.mrh]; render(); });\n chartsMRHist();\n }\n return;\n }\n if (section.id === \"marginal\") {\n const sd = $(\"#selMgDona\"); if (sd) sd.onchange = e => { S.mgDona = e.target.value; render(); };\n document.querySelectorAll(\"tr.grp[data-mgr]\").forEach(r => r.onclick = () => { S.mgExp[r.dataset.mgr] = !S.mgExp[r.dataset.mgr]; render(); });\n document.querySelectorAll(\"tr.child[data-mge]\").forEach(r => r.onclick = () => { S.mgExpE[r.dataset.mge] = !S.mgExpE[r.dataset.mge]; render(); });\n chartsMarginal();\n return;\n }\n if (section.id === \"factores-de-riesgo\") {\n document.querySelectorAll(\"[data-frr]\").forEach(b => b.onclick = () => { S.frRating = b.dataset.frr; render(); });\n document.querySelectorAll(\"tr.grp[data-frf]\").forEach(r => r.onclick = () => { S.frExp[r.dataset.frf] = !S.frExp[r.dataset.frf]; render(); });\n chartsFr();\n return;\n }\n if (section.id === \"backtesting\") {\n document.querySelectorAll(\"[data-btq]\").forEach(b => b.onclick = () => { S.btQ = b.dataset.btq; render(); });\n chartsBt();\n return;\n }\n if (section.id !== \"portafolio\") return;\n if (tab.id === \"global\") {\n const sd = $(\"#selDona\"); if (sd) sd.onchange = e => { S.donaVista = e.target.value; render(); };\n const sb = $(\"#selBL\"); if (sb) sb.onchange = e => { S.blVista = e.target.value; render(); };\n document.querySelectorAll(\"tr.grp\").forEach(r => r.onclick = () => { S.expanded[r.dataset.rat] = !S.expanded[r.dataset.rat]; render(); });\n chartsGlobal();\n } else {\n document.querySelectorAll(\"[data-tl]\").forEach(b => b.onclick = () => { S.concTL = b.dataset.tl; render(); });\n document.querySelectorAll(\"[data-tr]\").forEach(b => b.onclick = () => { S.concTR = b.dataset.tr; render(); });\n chartsConc();\n }\n }\n\n // ---------- Panel de filtros (orden inviolable del contrato) ----------\n function renderFilters() {\n const host = $(\"#filters\");\n if (!S.filtersOpen) { host.innerHTML = \"\"; return; }\n host.innerHTML = `

Filtros

Fecha de corte

Portafolio

Dataset

Calificación

Emisor

Tipo de instrumento

Seniority

Moneda\n ${[\"metricas-de-riesgo\", \"marginal\", \"factores-de-riesgo\"].includes(parseHash().section.id)\n ? `${[\"Todas\", \"DOP\", \"USD\"].map(m => ``).join(\"\")}\n

No aplica en esta sección: las medidas se calculan sobre la cartera consolidada completa (el ${parseHash().section.id === \"factores-de-riesgo\" ? \"VaR\" : S.medida} no es aditivo entre monedas${parseHash().section.id === \"marginal\" ? \"; el marginal leave-one-out se define contra la cartera completa\" : parseHash().section.id === \"factores-de-riesgo\" ? \"; la atribución por factor opera sobre el VaR consolidado y jamás se parte por moneda\" : \"\"}).

`\n : `${[\"Todas\", \"DOP\", \"USD\"].map(m => ``).join(\"\")}`}

Vista por\n

Dimensión de agregación (no es filtro) — sincronizada con el selector del gráfico de composición.

Dominios autoritativos: los valores válidos provienen de la capa de datos (mock: declarados del contrato backend). Controles no interactivos en esta iteración salvo Moneda y Vista por.

`;\n $(\"#fOverlay\").onclick = () => { S.filtersOpen = false; renderFilters(); };\n host.querySelectorAll(\"[data-mon]\").forEach(b => b.onclick = () => { S.moneda = b.dataset.mon; render(); S.filtersOpen = true; renderFilters(); });\n const fv = $(\"#fVista\"); if (fv) fv.onchange = e => { S.donaVista = e.target.value; render(); S.filtersOpen = true; renderFilters(); };\n }\n\n // ---------- Switcher de estado del prototipo ----------\n const STATES = [[\"ok\", \"Éxito\"], [\"loading\", \"Cargando\"], [\"empty\", \"Vacío\"], [\"error\", \"Error\"], [\"partial\", \"Error parcial\"]];\n function renderSwitch() {\n $(\"#protoSwitch\").innerHTML = `

Estado del prototipo

` +\n STATES.map(([k, l]) => ``).join(\"\") + `

\n \n

\n ${selMedida}\n ⓘ ${sup}\n ${chipVentana}\n ${dataBadge}\n ${D.DATASET} ▾\n \n

`;\n }\n function tabsHtml(section, tab) {\n if (section.noTabsBar || !section.tabs.length) return `

`;\n return `

` + section.tabs.map(t =>\n ``).join(\"\") + `

`;\n if (S.uiState === \"error\") return `

${errBox()}

`;\n if (S.uiState === \"empty\") return `

${emptyBox()}

`;\n return `

`;\n }\n\n const emptyBox = () => `

◌\n Sin datos para este corte\n El dataset seleccionado aún no tiene resultados. Selecciona otro dataset en el selector superior o ejecuta el cálculo del corte.

`;\n const errBox = () => `

⚠\n No se pudo cargar la información\n Ocurrió un error al consultar los datos del corte. Puedes reintentar la carga.\n

`;\n\n function kpiCard(title, big, sub, rows) {\n return `

${title}

${big}

\n ${sub ? `

${sub}

` : \"\"}\n ${rows && rows.length ? `

` + rows.map(r =>\n `

${r[0]}${r[1]}

`).join(\"\") + `

` : \"\"}\n

`;\n }\n function kpiSkeletons() { return `

`.repeat(5); }\n function panelSkeletons(n = 4) { return `

`.repeat(n); }\n function statePanels(box, n = 4) { return Array(n).fill(`

${box}

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisND(\"N/D\")}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisND(\"N/D\")}

${statePanels(errBox())}

${kpis}

Valor de Mercado por Calificación

Temporalidad: Año a la Fecha ▾ · Eje en DOP MM. Cada punto consolidado al TC de su fecha.

${S.uiState === \"partial\" ? errBox() : `\n

Composición del Portafolio

Participación sobre VM Total (DOP consolidado).${S.donaVista === \"Emisor\" ? \" Vista por emisor: top-7 + “Otros”.\" : \"\"}

Cambio en la Composición

Δ VM vs. corte anterior (DOP consolidado); barras en neutro — el signo lo comunica la posición y la etiqueta.

Tabla de Composición

${tablaGlobal()}

Variaciones al TC de cada corte; incluyen efecto FX. “—” = sin posiciones en el grupo; “N/D” = sin corte anterior (nunca 0).

Calificación	Valor de Mercado	% Global	Var (DOP)	Var %
DOP	USD	Total (DOP)
${open ? \"▾\" : \"▸\"}${r}	${cell(a.dop)}	${a.usd > 0 ? \"$\" + nf2.format(a.usd) + \"MM\" : \"—\"}	${fMM(a.tot)}	${fPct(a.tot / t * 100)}
${esc(e)} ${chip}	${cell(ea.dop)}	${ea.usd > 0 ? \"$\" + nf2.format(ea.usd) + \"MM\" : \"—\"}	${fMM(ea.tot)}	${fPct(ea.tot / t * 100)}

`;\n }\n\n // ====== Portafolio › Concentración (§7.2) ======\n function concHtml() {\n if (S.uiState === \"loading\") return `

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDConc(\"N/D\")}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDConc(\"N/D\")}

${statePanels(errBox())}

${kpis}

Top Emisores

Top-7 + “Otros” por ${S.concTL === \"pct\" ? \"% del VM\" : \"VM (DOP consolidado)\"}.

${S.uiState === \"partial\" ? errBox() : `\n

Evolución de la Concentración

Línea punteada = límite interno. Cada punto consolidado al TC de su fecha.

Cambio en la Composición por Emisor

Δ VM vs. corte anterior — top-7 por |Δ| + “Otros” + Total; barras en neutro.

Tabla de Concentración por Emisor

${tablaConc(rows, hhi)}

Lista completa (exenta de top-7): un emisor en alerta o exceso jamás queda oculto. Límites de exposición sobre VM observado. Sin límite → “N/D”.

Emisor	Moneda	VM (DOP)	% VM	Aporte HHI	Límite interno	Holgura/Exceso
${esc(r.emisor)}	${r.chip}	${fMM(r.vm)}	${fPct(r.pct)}	${r.hhi.toFixed(4)}	${fPct(D.LIMITES.emisorPctInterno, 1)}	${(holg >= 0 ? \"+\" : \"\") + nf1.format(holg)} pp ${st.txt}

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(MR_F_TITLES())}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(MR_F_TITLES())}

${statePanels(errBox())}

${kpis}

${S.medida} por Calificación

Contribución a la cartera por rating, orden de escala AAA→B (DOP MM). Cero real = barra en 0; sin dato = barra ausente + “N/D”.

${S.uiState === \"partial\" ? errBox() : `\n

Contribución al ${S.medida}

${S.mrDona === \"Emisor\" ? \"Top-7 + “Otros”. \" : \"\"}${LEYENDA_PRORRATEO}

Standalone vs. Contribución por Calificación

La brecha entre ${S.medida} standalone y contribución a cartera es la diversificación. La vista por posición vive en Marginal.

Descomposición del ${S.medida}

${tablaMrFecha(a)}

Solo el VM se parte por moneda; el ${S.medida} va siempre en DOP consolidado (no aditivo entre monedas). Filas de emisor: ${LEYENDA_PRORRATEO} Las variaciones viven en el tab Histórico.

${notaMonedaMR}

Calificación	Valor de Mercado	${S.medida} (DOP)	% del ${S.medida}
DOP	USD	Total (DOP)
${open ? \"▾\" : \"▸\"}${r}	${vmDop > 0 ? fMM(vmDop) : \"—\"}	${vmUsd > 0 ? \"$\" + nf2.format(vmUsd) + \"MM\" : \"—\"}	${fMM(vmTot)}	${fMM(contribR)}	${fPct(contribR / mrAgg().total * 100)}
${esc(e)} ${chip}	${eDop > 0 ? fMM(eDop) : \"—\"}	${eUsd > 0 ? \"$\" + nf2.format(eUsd) + \"MM\" : \"—\"}	${fMM(eTot)}	${fMM(eContrib)} prorr.	${fPct(eContrib / mrAgg().total * 100)}

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(MR_H_TITLES())}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(MR_H_TITLES())}

${statePanels(errBox())}

${kpis}

Evolución del ${S.medida} por Calificación

Cima del apilado = ${S.medida} Total. Línea vertical = cambio de supuesto — la serie no se lee como continua entre supuestos distintos (§2). Cada punto consolidado al TC de su fecha.

${S.uiState === \"partial\" ? errBox() : `\n

Evolución de la Razón ${S.medida}/VM

Intensidad de riesgo del libro (%). Mismos marcadores de cambio de supuesto que la serie de la izquierda.

Cambio del ${S.medida} vs. Corte Anterior

Comparativa por Calificación

${tablaMrHist(a, ap)}

${notaMonedaMR}

Calificación	Valor de Mercado	${S.medida} (DOP)	Var (DOP)	Var %
DOP	USD	Total (DOP)
${(cur.contrib - prev.contrib >= 0 ? \"+\" : \"−\") + fMM(Math.abs(cur.contrib - prev.contrib)).slice(1)}	${fPct((cur.contrib - prev.contrib) / prev.contrib * 100)}	N/D	N/D
${open ? \"▾\" : \"▸\"}${r}	${vmDop > 0 ? fMM(vmDop) : \"—\"}	${vmUsd > 0 ? \"$\" + nf2.format(vmUsd) + \"MM\" : \"—\"}	${fMM(vmTot)}	${fMM(cur.contrib)}
${(eContrib - eContribPrev >= 0 ? \"+\" : \"−\") + fMM(Math.abs(eContrib - eContribPrev)).slice(1)}	${fPct((eContrib - eContribPrev) / eContribPrev * 100)}	N/D	N/D
${esc(e)} ${chip}	${eDop > 0 ? fMM(eDop) : \"—\"}	${eUsd > 0 ? \"$\" + nf2.format(eUsd) + \"MM\" : \"—\"}	${fMM(eTot)}	${fMM(eContrib)} prorr.

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(MG_TITLES())}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(MG_TITLES())}

${statePanels(errBox())}

`;\n\n return `${glos}

${kpis}

Iso-risk: ${S.medida} marginal vs. tamaño

Cada punto = instrumento; X·Y = marginal en DOP. Overlays de límite conmutan con la medida; sin límite configurado → “Límite: N/D”. Diversificadoras bajo Y=0. ${NOTA_GRANULARIDAD}

${S.uiState === \"partial\" ? errBox() : `\n

Contribución (componental)

Ranking por ${S.medida} marginal

Top-7 posiciones por |marginal| mostrando signo: añade (+) / diversifica (−). “Otros (+)” y “Otros (−)” separados — un neto ocultaría magnitudes de ambos signos.

Detalle por posición

${tablaMg(rows, total)}

${notaMonedaMR}

Calificación / Emisor / Instrumento	Moneda	VM (DOP)	${S.medida} Marginal	Componental	% del total	Standalone	Benef. diversif.	Var Marginal	Var %
${open ? \"▾\" : \"▸\"}${r}	—	${fMM(vmR)}	—	${fSg(compR)}	${fPct(compR / total * 100)}	${fMM(standR)}	${fSg(standR - compR)}	—	—
${openE ? \"▾\" : \"▸\"}${esc(e)}	${chip}	${fMM(vmE)}	—	${fSg(compE)}	${fPct(compE / total * 100)}	${fMM(standE)}	${fSg(standE - compE)}	—	—
${esc(x.p.inst)}	${x.p.moneda}	${fMM(x.v)}	${fSg(x.marg)}	${fSg(x.comp)}	${fPct(x.comp / total * 100)}	${fMM(x.stand)}	${fSg(x.stand - x.comp)}	${dv === null ? \"N/D\" : (dv >= 0 ? \"+\" : \"−\") + fMM(Math.abs(dv)).slice(1)}	${dvp === null ? \"N/D\" : (dvp >= 0 ? \"+\" : \"\") + nf1.format(dvp) + \"%\"}

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDMR(BT_TITLES)}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDMR(BT_TITLES)}

${statePanels(errBox())}

${kpis}

Serie de excepciones ⓘ

${BT_CONV}.

Series en neutro; la excepción se marca por forma (rombo) en un único color de alerta rotulado — nunca la terna semafórica. Cada punto consolidado al TC de su fecha.

${S.uiState === \"partial\" ? errBox() : btTRHtml(w)}

Convergencia del estimador MC ⓘ

\n ${[\"1%\", \"0.5%\", \"0.1%\"].map(q =>\n ``).join(\"\")}

Colas profundas: la banda de confianza escala ~1/q en escenarios.
Bandas del estimador Monte Carlo — no es la distribución de valor del portafolio.

Cuantil de diagnóstico: ${S.btQ} · eje Y fijo común entre cuantiles, a escala real — el ensanchamiento de la banda es el caveat.

Excepciones del VaR — 99% · 1 año (DOP consolidado)

${tablaBt(w)}

`;\n }\n function btTRHtml(w) {\n const pv = D.BT.prevWin;\n return `

Validación estadística ⓘ

Ventana válida: ${w.n} cortes desde ${w.val[0].f} (objetivo 250)

Kupiec POF

${ND_WIN(w.n)}

LR: N/D · χ²(1) · p-value: N/D · veredicto: N/D

Christoffersen (conjunto: independencia + cobertura)

${ND_WIN(w.n)}

LR conjunto: N/D · χ²(2) · p-value: N/D · veredicto: N/D

Semáforo de Basilea

${ND_WIN(w.n)}${BT_ROT_BASILEA}

Zonas normalizadas a 250 cortes: verde 0–4 · amarilla 5–9 · roja 10+ · multiplicador base 3 + recargo (referencia ilustrativa del marco de mercado)

Ventana anterior (cerrada) — referencia histórica, no es el veredicto vigente

${pv.rot} · ${pv.exc} excepciones

\n Kupiec POF${pv.kupiec.v}LR ${pv.kupiec.lr} · χ²(1) · p=${pv.kupiec.p}\n Christoffersen (conjunto)${pv.christo.v}LR ${pv.christo.lr} · χ²(2) · p=${pv.christo.p}\n

Zona ${pv.basilea.zona} · ${pv.basilea.conteo}${BT_ROT_BASILEA}

Multiplicador ${pv.basilea.mult} (referencia ilustrativa del marco de mercado)

Excepciones contadas bajo la convención de escala del mock. Las pruebas se computan solo sobre la sub-ventana homogénea vigente — jamás cruzan un cambio de supuesto (§2).

Fecha	P&L realizado (DOP MM)	VaR 99%·1a escalado (DOP MM)	Exceso (DOP MM)	Supuesto vigente	Ventana válida

${kpiSkeletons()}

${panelSkeletons()}

`;\n if (S.uiState === \"empty\") return `

${kpisNDFR()}

${statePanels(emptyBox())}

`;\n if (S.uiState === \"error\") return `

${kpisNDFR()}

${statePanels(errBox())}

${kpis}

Contribución por factor — Atribución Euler (VaR, DOP MM)

${frLeyendaTL(a)} El “Efecto correlación” es el marcador anotado — nunca una barra. ${FR_CONV}.

${S.uiState === \"partial\" ? errBox() : `\n

Umbrales de Merton — variable latente

\n ${[\"AAA\", \"AA\", \"A\", \"BBB\", \"BB\", \"B\", \"CCC\"].map(r =>\n ``).join(\"\")}

${FR_GAUSS} — la silueta multimodal del valor vive en Escenarios.

Matriz ρ asumida (correlación de activos)

Contribución al VaR — 99% · 1 año (DOP consolidado)

${tablaFr(a, ap)}

Variaciones al TC de cada corte; incluyen efecto FX — la contribución va en DOP consolidado y jamás se parte por moneda. Totales calculados sobre valores sin redondear. ${FR_CONV}.

Factor / Calificación	Contribución (DOP MM)	% del VaR	Var (DOP MM)	Var %

Filtros

Fecha de corte

Portafolio

Dataset

Calificación

Emisor

Tipo de instrumento

Seniority

Moneda\n ${[\"metricas-de-riesgo\", \"marginal\", \"factores-de-riesgo\"].includes(parseHash().section.id)\n ? `${[\"Todas\", \"DOP\", \"USD\"].map(m => ``).join(\"\")}\n

`\n : `${[\"Todas\", \"DOP\", \"USD\"].map(m => ``).join(\"\")}`}

Vista por\n

Dimensión de agregación (no es filtro) — sincronizada con el selector del gráfico de composición.

Dominios autoritativos: los valores válidos provienen de la capa de datos (mock: declarados del contrato backend). Controles no interactivos en esta iteración salvo Moneda y Vista por.

Estado del prototipo

` +\n STATES.map(([k, l]) => ``).join(\"\") + `

`;\n document.querySelectorAll(\"[data-st]\").forEach(b => b.onclick = () => { S.uiState = b.dataset.st; render(); });\n }\n\n window.addEventListener(\"hashchange\", render);\n window.addEventListener(\"resize\", () => charts.forEach(c => c.resize()));\n if (!location.hash) location.hash = \"#/vision-general\";\n render();\n})();\n","createdAt",["D",1787332301862],"branchVersion","branchId","26318b20-79c4-45b1-8b64-9b218f217a44","id","935f65cd-5733-4dbc-b0ca-b5a3502394d7","ri-credito-mock",["D",1787326658800],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","Mock UX/UI seccion Credito de Risk Intelligence (F2.2)","imageUrl","019c346e-384f-704b-b25a-da5eb72ef468","pinnedFileId","64f125fa-9d76-11f1-b549-1607ee4eb77e","pinPreview",true,"pinMigrated","screenshotVersion","likeCount",0,"orgId","019c346e-384f-704b-b25a-dc67b1d14ed5","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_97":96,"_104":100},"org",{"_97":96,"_98":-5,"_99":100,"_18":55,"_101":102,"_103":25},"branch",{"_13":86,"_32":87,"_43":42,"_88":44,"_89":-5,"_90":-5,"_91":-5,"_92":-5,"_93":-5,"_94":45,"_95":96},"isOwned",110,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","joselier3","handle","displayName","avatar","https://img.clerk.com/eyJ0eXBlIjoicHJveHkiLCJzcmMiOiJodHRwczovL2ltYWdlcy5jbGVyay5kZXYvb2F1dGhfZ2l0aHViL2ltZ18zOUpFemx0ZkZMYnQwNks1d1VoUmR5blBQelUifQ","tier","free","canRestrictVals","profileImageUrl","branchName","","adoptOrgId",{"_43":44,"_32":45,"_38":201,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":57,"_58":59,"_60":-5,"_61":-5,"_62":63,"_64":65,"_66":-5,"_67":-5,"_68":-5,"_69":-5,"_70":71,"_72":71,"_73":74,"_75":74,"_76":74,"_77":71,"_78":71,"_79":202,"_81":203,"_83":204,"_85":25},"tree",[180,181,182,183,184],"files",[180,181,182,183,184],"counts",{"_170":63,"_171":155,"_172":63,"_173":174,"_175":176},"contents",[127,128,129,130,131],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_43":166,"_32":28,"_30":31,"_135":-5,"_41":42,"_40":14,"_137":25,"_38":167,"_13":168,"_33":34,"_35":37,"_24":25,"_141":-5,"_27":169},{"_43":158,"_32":159,"_30":160,"_135":-5,"_41":42,"_40":161,"_137":25,"_38":162,"_13":163,"_33":34,"_35":164,"_24":25,"_141":-5,"_27":165},{"_43":150,"_32":151,"_30":152,"_135":-5,"_41":42,"_40":153,"_137":59,"_38":154,"_13":155,"_33":20,"_35":156,"_24":25,"_141":-5,"_27":157},{"_43":143,"_32":144,"_30":57,"_135":-5,"_41":42,"_40":145,"_137":59,"_38":146,"_13":63,"_33":147,"_35":148,"_24":25,"_141":-5,"_27":149},{"_43":132,"_32":133,"_30":134,"_135":-5,"_41":42,"_40":136,"_137":25,"_38":138,"_13":139,"_33":20,"_35":140,"_24":25,"_141":-5,"_27":142},"16e8a70d-5256-484d-a81f-447e3ccf8817","styles.css","4ee4523c-9d76-11f1-b524-1607ee4eb77e","parentId",66,"latest",["D",1787331497474],3,"/* Risk Intelligence — Crédito (mock) — tokens del baseline RI (04-design-system-mapping.md)\n Prohibido hex fuera de :root [inviolable L-7] */\n:root {\n --bg-app: #eff3f8;\n --primary: #0d77bf;\n --secondary: #1f2a4e;\n --text: #000;\n --text-2: #434343;\n --border: #dfdfdf;\n --hover: #d9e2ed;\n --skeleton: #d9e2ed;\n --sem-error: #c10007;\n --fondo-error: #ffe2e2;\n --sem-ok: #008236;\n --fondo-ok: #dcfce7;\n --sem-warn: #de560d;\n --fondo-alerta: #ffedd4;\n --surface-container: #ffffff;\n --surface-card: #eff3f8;\n --radius-container: 20px;\n --radius-panel: 10px;\n --radius-card: 8px;\n --sp: 4px;\n --font: 'Inter', sans-serif;\n}\n* { box-sizing: border-box; margin: 0; padding: 0; }\nhtml, body { height: 100%; }\nbody { font-family: var(--font); background: var(--bg-app); color: var(--text); font-size: 13px; }\n.num { font-variant-numeric: tabular-nums; }\n\n/* ---------- Header global ---------- */\n.app-header {\n display: flex; align-items: center; gap: 16px;\n padding: 10px 24px; background: var(--bg-app);\n}\n.brand { font-weight: 800; font-size: 18px; color: var(--secondary); letter-spacing: 0.5px; }\n.brand small { display: block; font-size: 8px; font-weight: 600; letter-spacing: 2px; color: var(--text-2); }\n.tenant {\n display: flex; align-items: center; gap: 8px; background: var(--surface-container);\n border: 1px solid var(--border); border-radius: var(--radius-card); padding: 8px 12px;\n font-weight: 600; font-size: 13px;\n}\n.modules {\n margin: 0 auto; display: flex; gap: 4px; background: var(--surface-container);\n border-radius: 16px; padding: 6px; box-shadow: 0 1px 3px rgba(31,42,78,0.08);\n}\n.module-tab {\n display: flex; flex-direction: column; align-items: center; gap: 2px;\n padding: 8px 22px; border-radius: 12px; font-size: 11px; font-weight: 600;\n color: var(--text-2); border: none; background: transparent; cursor: default; font-family: var(--font);\n}\n.module-tab .mico { font-size: 15px; }\n.module-tab.active { background: var(--primary); color: #fff; }\n.module-tab.disabled { color: var(--border); }\n.header-right { display: flex; align-items: center; gap: 14px; }\n.avatar {\n width: 36px; height: 36px; border-radius: var(--radius-card); background: var(--secondary);\n color: #fff; display: flex; align-items: center; justify-content: center; font-weight: 700; font-size: 12px;\n}\n.hicon { color: var(--secondary); font-size: 16px; }\n\n/* ---------- Layout ---------- */\n.app-body { display: flex; align-items: flex-start; }\n.sidebar { width: 232px; flex-shrink: 0; padding: 8px 12px 24px 24px; }\n.sidebar h1 { font-size: 26px; font-weight: 800; color: var(--secondary); margin: 6px 0 18px 6px; }\n.menu-item {\n display: flex; align-items: center; gap: 10px; width: 100%;\n padding: 11px 14px; margin-bottom: 2px; border: none; background: transparent;\n border-radius: var(--radius-panel); font-family: var(--font); font-size: 13.5px; font-weight: 500;\n color: var(--text); cursor: pointer; text-align: left;\n}\n.menu-item:hover { background: var(--hover); }\n.menu-item.active { background: var(--surface-container); font-weight: 600; box-shadow: 0 1px 3px rgba(31,42,78,0.10); }\n.menu-item .ico { width: 18px; text-align: center; color: var(--secondary); }\n.main {\n flex: 1; background: var(--surface-container); border-radius: var(--radius-container);\n margin: 0 24px 24px 0; padding: 20px 24px 28px; min-height: 86vh; min-width: 0;\n}\n\n/* ---------- Cabecera de sección (badges FUERA de paneles skeletonizables — X-2) ---------- */\n.section-head { display: flex; align-items: flex-start; justify-content: space-between; gap: 16px; flex-wrap: wrap; }\n.breadcrumb { display: flex; align-items: center; gap: 8px; font-size: 14px; padding-top: 6px; }\n.breadcrumb .home { color: var(--secondary); }\n.breadcrumb .sep { color: var(--text-2); }\n.breadcrumb .leaf { font-weight: 600; }\n.head-badges { display: flex; align-items: center; gap: 10px; flex-wrap: wrap; justify-content: flex-end; }\n.badge-supuesto {\n display: inline-flex; align-items: center; gap: 6px;\n background: var(--surface-card); border: 1px solid var(--border);\n border-radius: 999px; padding: 7px 14px; font-size: 12px; color: var(--text-2); font-weight: 500;\n}\n.badge-supuesto .lbl { font-weight: 700; color: var(--secondary); }\n.badge-data {\n display: inline-flex; align-items: center; gap: 8px; background: var(--surface-card);\n border-radius: var(--radius-card); padding: 8px 14px; font-size: 13px;\n}\n.badge-data b { font-weight: 700; }\n.badge-data.warn { background: var(--fondo-alerta); color: var(--sem-warn); }\n.badge-data.err { background: var(--fondo-error); color: var(--sem-error); }\n.dataset-select, .btn-filtros {\n display: inline-flex; align-items: center; gap: 8px; background: var(--surface-container);\n border: 1px solid var(--border); border-radius: var(--radius-card); padding: 8px 14px;\n font-family: var(--font); font-size: 13px; cursor: pointer;\n}\n.btn-filtros { font-weight: 600; }\n.btn-filtros .count {\n background: var(--primary); color: #fff; border-radius: 999px;\n font-size: 10px; width: 16px; height: 16px; display: inline-flex; align-items: center; justify-content: center;\n}\n\n/* ---------- Tabs pill ---------- */\n.tabs-bar {\n display: inline-flex; gap: 6px; background: var(--surface-card);\n border-radius: 12px; padding: 6px; margin: 16px 0 18px;\n}\n.tab-pill {\n border: none; background: transparent; font-family: var(--font); font-size: 13px; font-weight: 600;\n color: var(--text-2); padding: 8px 20px; border-radius: var(--radius-card); cursor: pointer;\n}\n.tab-pill.active { background: var(--surface-container); color: var(--text); border: 1px solid var(--border); }\n\n/* ---------- Dashboard grid ---------- */\n.dash { display: grid; grid-template-columns: 230px 1fr; gap: 16px; align-items: start; }\n.kpi-col { display: flex; flex-direction: column; gap: 12px; }\n.kpi-card { background: var(--surface-card); border-radius: var(--radius-card); padding: 14px 16px; }\n.kpi-card h3 { font-size: 13px; font-weight: 700; margin-bottom: 6px; }\n.kpi-card .big { font-size: 21px; font-weight: 800; letter-spacing: -0.3px; }\n.kpi-card .sub { font-size: 10.5px; color: var(--text-2); margin-top: 3px; line-height: 1.35; }\n.kpi-card .rows { margin-top: 10px; display: flex; flex-direction: column; gap: 4px; }\n.kpi-card .row { display: flex; justify-content: space-between; font-size: 11.5px; color: var(--text-2); }\n.kpi-card .row b { color: var(--text); font-weight: 600; }\n.trend { font-weight: 700; color: var(--secondary); } /* Δ de riesgo/mezcla: neutro SIEMPRE (L-5/L-6) */\n.grid-2x2 { display: grid; grid-template-columns: 1fr 1fr; gap: 16px; min-width: 0; }\n.panel {\n background: var(--surface-container); border: 1px solid var(--border);\n border-radius: var(--radius-panel); padding: 14px 16px; min-width: 0; min-height: 300px;\n display: flex; flex-direction: column;\n}\n.panel-head { display: flex; align-items: center; justify-content: space-between; gap: 8px; margin-bottom: 8px; flex-wrap: wrap; }\n.panel-head h2 { font-size: 14.5px; font-weight: 700; }\n.chart-box { flex: 1; min-height: 230px; }\n.panel-foot { font-size: 10.5px; color: var(--text-2); margin-top: 8px; line-height: 1.4; }\n.seg {\n display: inline-flex; background: var(--surface-card); border-radius: var(--radius-card); padding: 3px; gap: 2px;\n}\n.seg button {\n border: none; background: transparent; font-family: var(--font); font-size: 11px; font-weight: 600;\n color: var(--text-2); padding: 5px 10px; border-radius: 6px; cursor: pointer;\n}\n.seg button.active { background: var(--surface-container); color: var(--text); border: 1px solid var(--border); }\n.select-mini {\n font-family: var(--font); font-size: 12px; border: 1px solid var(--border);\n border-radius: var(--radius-card); padding: 5px 8px; background: var(--surface-container);\n}\n\n/* ---------- Estado / pills ---------- */\n.pill { display: inline-flex; align-items: center; gap: 5px; border-radius: 999px; padding: 3px 10px; font-size: 11px; font-weight: 700; }\n.pill.ok { background: var(--fondo-ok); color: var(--sem-ok); }\n.pill.warn { background: var(--fondo-alerta); color: var(--sem-warn); }\n.pill.err { background: var(--fondo-error); color: var(--sem-error); }\n.pill.na { background: var(--surface-card); color: var(--text-2); border: 1px solid var(--border); }\n.chip-mon { display: inline-block; border: 1px solid var(--border); background: var(--surface-card); color: var(--text-2); border-radius: 4px; padding: 1px 7px; font-size: 10.5px; font-weight: 600; }\n\n/* ---------- Tabla ---------- */\n.tbl-wrap { flex: 1; overflow: auto; }\ntable.tbl { width: 100%; border-collapse: collapse; font-size: 12px; }\n.tbl th { text-align: right; font-weight: 700; padding: 7px 8px; border-bottom: 1px solid var(--border); white-space: nowrap; }\n.tbl th.gh { text-align: center; border-bottom: 1px solid var(--border); color: var(--text-2); font-size: 11px; }\n.tbl th:first-child, .tbl td:first-child { text-align: left; }\n.tbl td { padding: 7px 8px; border-bottom: 1px solid var(--surface-card); text-align: right; white-space: nowrap; }\n.tbl tr.grp td { font-weight: 700; cursor: pointer; }\n.tbl tr.grp:hover td { background: var(--hover); }\n.tbl tr.child td:first-child { padding-left: 26px; color: var(--text-2); }\n.tbl .caret { display: inline-block; width: 14px; color: var(--text-2); font-size: 10px; }\n.btn-export {\n border: 1px solid var(--border); background: var(--surface-container); border-radius: var(--radius-card);\n font-family: var(--font); font-size: 11.5px; font-weight: 600; padding: 5px 12px; cursor: pointer; color: var(--secondary);\n}\n\n/* ---------- Estados UI ---------- */\n.skeleton { border-radius: var(--radius-panel); background: var(--skeleton); animation: pulse 1.4s ease-in-out infinite; min-height: 300px; }\n.skeleton.sm { min-height: 96px; border-radius: var(--radius-card); }\n@keyframes pulse { 0%,100% { opacity: 1; } 50% { opacity: 0.55; } }\n@media (prefers-reduced-motion: reduce) { .skeleton { animation: none; } }\n.state-box { flex: 1; display: flex; flex-direction: column; align-items: center; justify-content: center; gap: 8px; text-align: center; padding: 20px; }\n.state-box .t { font-weight: 700; font-size: 13.5px; }\n.state-box .d { color: var(--text-2); font-size: 12px; max-width: 340px; line-height: 1.45; }\n.state-box .ic { font-size: 26px; }\n.btn-retry {\n margin-top: 6px; border: none; background: var(--primary); color: #fff; font-family: var(--font);\n font-weight: 600; font-size: 12px; padding: 8px 18px; border-radius: var(--radius-card); cursor: pointer;\n}\n\n/* ---------- Placeholder de sección ---------- */\n.placeholder-note {\n background: var(--surface-card); border: 1px dashed var(--border); border-radius: var(--radius-panel);\n padding: 34px; text-align: center; color: var(--text-2); margin-top: 8px;\n}\n.placeholder-note b { color: var(--secondary); display: block; font-size: 15px; margin-bottom: 6px; }\n\n/* ---------- Panel de filtros (derecho) ---------- */\n.filters-overlay { position: fixed; inset: 0; background: rgba(31,42,78,0.25); z-index: 40; }\n.filters-panel {\n position: fixed; top: 0; right: 0; bottom: 0; width: 320px; background: var(--surface-container);\n z-index: 41; padding: 20px; overflow-y: auto; box-shadow: -4px 0 16px rgba(31,42,78,0.15);\n}\n.filters-panel h2 { font-size: 16px; font-weight: 800; color: var(--secondary); margin-bottom: 14px; }\n.f-group { margin-bottom: 16px; }\n.f-group label { display: block; font-size: 11px; font-weight: 700; color: var(--text-2); margin-bottom: 5px; text-transform: uppercase; letter-spacing: 0.4px; }\n.f-group select, .f-group input {\n width: 100%; font-family: var(--font); font-size: 12.5px; border: 1px solid var(--border);\n border-radius: var(--radius-card); padding: 8px 10px; background: var(--surface-container);\n}\n.f-sep { border-top: 1px solid var(--border); margin: 14px 0; }\n.f-note { font-size: 10.5px; color: var(--text-2); line-height: 1.4; }\n\n/* ---------- Switcher de estado del prototipo ---------- */\n.proto-switch {\n position: fixed; bottom: 16px; right: 16px; z-index: 50;\n background: var(--secondary); color: #fff; border-radius: var(--radius-panel);\n padding: 10px 12px; box-shadow: 0 4px 14px rgba(31,42,78,0.35); font-size: 11px;\n}\n.proto-switch .t { font-weight: 700; margin-bottom: 6px; opacity: 0.8; }\n.proto-switch .opts { display: flex; gap: 4px; flex-wrap: wrap; }\n.proto-switch button {\n border: 1px solid rgba(255,255,255,0.3); background: transparent; color: #fff;\n font-family: var(--font); font-size: 10.5px; font-weight: 600; padding: 4px 9px; border-radius: 6px; cursor: pointer;\n}\n.proto-switch button.active { background: #fff; color: var(--secondary); }\n\n@media (max-width: 1100px) {\n .dash { grid-template-columns: 1fr; }\n .kpi-col { flex-direction: row; flex-wrap: wrap; }\n .kpi-card { flex: 1 1 180px; }\n .grid-2x2 { grid-template-columns: 1fr; }\n}\n\n/* F2.3 — Métricas de Riesgo */\n.seg-medida { margin-right: 6px; }\n.seg-medida button { font-weight: 600; min-width: 40px; }\n.seg button[disabled] { opacity: .45; cursor: not-allowed; }\n.mr-nota { grid-column: 1 / -1; margin-top: 10px; padding: 8px 12px; font-size: 11px; color: var(--text-2); background: var(--card-bg); border: 1px dashed var(--border); border-radius: var(--radius); }\n\n/* F2.10 — Backtesting */\n.chip-window { display: inline-flex; align-items: center; border: 1px solid var(--border); background: var(--surface-card); color: var(--text-2); border-radius: 999px; padding: 7px 12px; font-size: 11.5px; font-weight: 600; }\n.bt-val { flex: 1; display: flex; flex-direction: column; gap: 10px; overflow: auto; }\n.bt-winhead { font-size: 12px; font-weight: 700; color: var(--secondary); background: var(--surface-card); border-radius: var(--radius-card); padding: 7px 10px; }\n.bt-block { border: 1px solid var(--border); border-radius: var(--radius-card); padding: 9px 11px; }\n.bt-block h4 { font-size: 12px; font-weight: 700; margin-bottom: 4px; }\n.bt-nd { font-size: 13px; font-weight: 700; color: var(--text-2); }\n.bt-det { font-size: 10.5px; color: var(--text-2); margin-top: 4px; line-height: 1.4; }\n.badge-basilea { display: block; border-radius: 6px; border: 1px solid var(--border); border-left: 4px solid var(--text-2); background: var(--surface-card); padding: 7px 10px; margin-top: 4px; }\n.badge-basilea .z { display: block; font-size: 12.5px; font-weight: 800; }\n.badge-basilea .r { display: block; font-size: 9.5px; color: var(--text-2); margin-top: 2px; font-weight: 600; text-transform: uppercase; letter-spacing: 0.3px; }\n.badge-basilea.arch { border-left-color: var(--sem-ok); background: var(--fondo-ok); }\n.badge-basilea.arch .z { color: var(--sem-ok); }\n.bt-archived { border: 1px dashed var(--border); border-radius: var(--radius-card); padding: 9px 11px; opacity: 0.78; }\n.bt-arch-rot { font-size: 10px; font-weight: 800; text-transform: uppercase; letter-spacing: 0.4px; color: var(--text-2); margin-bottom: 4px; }\n.bt-arch-sub { font-size: 11px; color: var(--text-2); margin-bottom: 6px; }\n.bt-arch-grid { display: grid; grid-template-columns: auto auto 1fr; gap: 3px 10px; font-size: 11.5px; align-items: baseline; }\n.bt-arch-grid b { font-weight: 700; }\n.bt-arch-grid .num { color: var(--text-2); font-size: 10.5px; }\n.bt-sub2 { font-size: 10.5px; color: var(--text-2); line-height: 1.4; margin: -2px 0 6px; }\n.bt-sub2 b { font-weight: 700; }\n.tbl tr.bt-out td { color: var(--text-2); opacity: 0.62; }\n\n/* F2.8 — Factores de Riesgo: subtítulo anti-Gauss del panel TR (cuerpo capturable, no tooltip — X-#2) */\n.fr-sub { font-size: 11px; font-style: italic; color: var(--text-2); padding: 4px 14px 0; }\n","label",[133],"291d60c8-c08e-40ea-b565-244e643c6f2d","main.ts",5,["D",1787326725014],"http","import { readFile } from \"https://esm.town/v/std/utils/index.ts\";\n\nexport default async function (req: Request): Promise {\n const path = new URL(req.url).pathname;\n const serve = async (file: string, type: string) =>\n new Response(await readFile(file, import.meta.url), {\n headers: { \"Content-Type\": type + \"; charset=utf-8\" },\n });\n if (path.endsWith(\"styles.css\")) return serve(\"/styles.css\", \"text/css\");\n if (path.endsWith(\"data.js\")) return serve(\"/data.js\", \"text/javascript\");\n if (path.endsWith(\"app.js\")) return serve(\"/app.js\", \"text/javascript\");\n return serve(\"/index.html\", \"text/html\");\n}\n",[144],"ce72ce6a-1f5e-4feb-9ed1-08216645377e","index.html","610320f6-9d76-11f1-8101-1607ee4eb77e",14,["D",1787327075110],1,"\n\n\n \n \n Crédito — Risk Intelligence (mock)\n \n \n \n \n\n\n \n

\n \n

\n \n \n\n\n",[151],"9b95084e-2889-4139-9be4-408b873944ed","data.js","5d60d72c-9d76-11f1-bcc0-1607ee4eb77e",69,["D",1787332294375],11,"/* Datos mock — deuda dominicana (bonos, SBB, SBB c/ Garantía, CDs), DOP + USD.\n Convención del mock: TC 61.50 (BCRD venta) al as_of 30/01/2026; cortes históricos\n consolidados al TC de su fecha. VM en MM DOP (o MM USD nativo cuando moneda=USD). */\nwindow.RI_DATA = (() => {\n const TC = 61.50, TC_PREV = 61.32;\n /* inst = nombre de instrumento (nivel hoja, export estable). mgF = factor marginal/componental\n por posición (mock determinista, 1.02–1.10: garantiza componental ≤ marginal ≤ standalone en\n contribuidoras). div = diversificadora: componental negativa por la REALIZACIÓN SIMULADA de su\n escenario (consistente con la nota de granularidad §2 — no es atributo del emisor); el excedente\n se redistribuye intra-rating para preservar Σ componental = total de Métricas de Riesgo. */\n const positions = [\n { inst: \"Bono Hacienda 2031\", emisor: \"Ministerio de Hacienda\", rating: \"AA\", tipo: \"Bono\", moneda: \"DOP\", vm: 612.40, prev: 605.10, mgF: 1.07 },\n { inst: \"Bono Global 2033\", emisor: \"Ministerio de Hacienda\", rating: \"AA\", tipo: \"Bono\", moneda: \"USD\", vm: 6.90, prev: 6.90, mgF: 1.09 },\n { inst: \"Bono Hacienda 2028\", emisor: \"Ministerio de Hacienda\", rating: \"AA\", tipo: \"Bono\", moneda: \"DOP\", vm: 291.00, prev: 296.40, mgF: 1.04 },\n { inst: \"Nota BCRD 2030\", emisor: \"Banco Central (BCRD)\", rating: \"AAA\", tipo: \"Bono\", moneda: \"DOP\", vm: 734.20, prev: 741.80, mgF: 1.05 },\n { inst: \"CD BCRD 91d\", emisor: \"Banco Central (BCRD)\", rating: \"AAA\", tipo: \"CD\", moneda: \"DOP\", vm: 210.00, prev: 210.00, mgF: 1.02 },\n { inst: \"Bono Banreservas 2029\", emisor: \"Banreservas\", rating: \"A\", tipo: \"Bono\", moneda: \"DOP\", vm: 498.60, prev: 502.30, mgF: 1.08 },\n { inst: \"CD Banreservas 180d\", emisor: \"Banreservas\", rating: \"A\", tipo: \"CD\", moneda: \"DOP\", vm: 187.50, prev: 172.00, mgF: 1.03 },\n { inst: \"Bono Popular 2027\", emisor: \"Banco Popular\", rating: \"A\", tipo: \"Bono\", moneda: \"DOP\", vm: 402.10, prev: 398.90, mgF: 1.06 },\n { inst: \"SBB Popular 2026\", emisor: \"Banco Popular\", rating: \"A\", tipo: \"SBB\", moneda: \"USD\", vm: 3.20, prev: 3.20, mgF: 1.05 },\n { inst: \"SBB BHD 2027\", emisor: \"Banco BHD\", rating: \"A\", tipo: \"SBB\", moneda: \"DOP\", vm: 356.80, prev: 361.20, mgF: 1.07 },\n { inst: \"CD BHD 90d\", emisor: \"Banco BHD\", rating: \"A\", tipo: \"CD\", moneda: \"DOP\", vm: 122.00, prev: 122.00, mgF: 1.00, div: true },\n { inst: \"SBB Gar. APAP 2028\", emisor: \"APAP\", rating: \"BBB\", tipo: \"SBB c/ Garantía\", moneda: \"DOP\", vm: 310.40, prev: 318.70, mgF: 1.08 },\n { inst: \"CD APAP 120d\", emisor: \"APAP\", rating: \"BBB\", tipo: \"CD\", moneda: \"DOP\", vm: 96.30, prev: 96.30, mgF: 1.02 },\n { inst: \"SBB Santa Cruz 2026\", emisor: \"Banco Santa Cruz\", rating: \"BBB\", tipo: \"SBB\", moneda: \"DOP\", vm: 228.90, prev: 224.10, mgF: 1.05 },\n { inst: \"SBB Gar. ACAP 2027\", emisor: \"ACAP\", rating: \"BBB\", tipo: \"SBB c/ Garantía\", moneda: \"DOP\", vm: 204.50, prev: 209.80, mgF: 1.04 },\n { inst: \"Bono CEPM 2030\", emisor: \"CEPM\", rating: \"BBB\", tipo: \"Bono\", moneda: \"USD\", vm: 5.40, prev: 5.10, mgF: 1.09 },\n { inst: \"Bono EGE Haina 2028\", emisor: \"EGE Haina\", rating: \"BB\", tipo: \"Bono\", moneda: \"DOP\", vm: 246.70, prev: 251.30, mgF: 1.08 },\n { inst: \"Bono EGE Haina 2031\", emisor: \"EGE Haina\", rating: \"BB\", tipo: \"Bono\", moneda: \"USD\", vm: 2.10, prev: 2.10, mgF: 1.06 },\n { inst: \"Bono AES Andrés 2029\", emisor: \"AES Andrés\", rating: \"BB\", tipo: \"Bono\", moneda: \"USD\", vm: 4.60, prev: 4.85, mgF: 1.10 },\n { inst: \"CD Promerica 180d\", emisor: \"Banco Promerica\", rating: \"BB\", tipo: \"CD\", moneda: \"DOP\", vm: 138.20, prev: 138.20, mgF: 1.00, div: true },\n { inst: \"SBB Caribe 2026\", emisor: \"Banco Caribe\", rating: \"B\", tipo: \"SBB\", moneda: \"DOP\", vm: 92.40, prev: 98.60, mgF: 1.09 },\n { inst: \"CD Caribe 90d\", emisor: \"Banco Caribe\", rating: \"B\", tipo: \"CD\", moneda: \"DOP\", vm: 41.00, prev: 41.00, mgF: 1.02 },\n ];\n const RATINGS = [\"AAA\", \"AA\", \"A\", \"BBB\", \"BB\", \"B\"];\n const TIPOS = [\"Bono\", \"SBB\", \"SBB c/ Garantía\", \"CD\"];\n const DATES = [\"04/12/2025\",\"09/12/2025\",\"14/12/2025\",\"19/12/2025\",\"24/12/2025\",\"29/12/2025\",\"03/01/2026\",\"08/01/2026\",\"13/01/2026\",\"18/01/2026\",\"20/01/2026\",\"24/01/2026\",\"28/01/2026\",\"30/01/2026\"];\n const HIST_TOTAL = [6020,6035,6018,6060,6072,6058,6094,6110,6102,6125,6139,6131,6146,0];\n const HIST_SHARE = {\n AAA: [0.20,0.20,0.20,0.20,0.20,0.19,0.19,0.19,0.19,0.19,0.19,0.18,0.18,0],\n AA: [0.21,0.21,0.21,0.21,0.21,0.21,0.21,0.21,0.21,0.22,0.22,0.22,0.22,0],\n A: [0.28,0.28,0.28,0.29,0.29,0.29,0.29,0.29,0.30,0.30,0.30,0.30,0.30,0],\n BBB: [0.17,0.17,0.17,0.16,0.16,0.17,0.17,0.17,0.16,0.16,0.16,0.17,0.17,0],\n BB: [0.11,0.11,0.11,0.11,0.11,0.11,0.11,0.11,0.11,0.10,0.10,0.10,0.10,0],\n B: [0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0],\n };\n const LIMITES = { hhiInterno: 0.1800, emisorPctInterno: 15.0, alertaFactor: 0.9 };\n /* Riesgo por medida (mock): tasa standalone (% del VM) y factor de diversificación,\n AMBOS por rating y por medida (X-2: proporciones distintas por medida — ES/VaR crece\n hacia ratings bajos, colas más pesadas). divF≤1 en todas (X-1: ES y σ son subaditivas;\n el beneficio de diversificación del mock es siempre ≥0). */\n const RISK = {\n VaR: { rate: { AAA: 0.0040, AA: 0.0090, A: 0.0180, BBB: 0.0360, BB: 0.0700, B: 0.1250 },\n divF: { AAA: 0.88, AA: 0.85, A: 0.80, BBB: 0.76, BB: 0.72, B: 0.70 } },\n ES: { rate: { AAA: 0.0046, AA: 0.0108, A: 0.0234, BBB: 0.0504, BB: 0.1050, B: 0.2000 },\n divF: { AAA: 0.90, AA: 0.87, A: 0.82, BBB: 0.78, BB: 0.74, B: 0.72 } },\n Std: { rate: { AAA: 0.0016, AA: 0.0034, A: 0.0066, BBB: 0.0126, BB: 0.0235, B: 0.0410 },\n divF: { AAA: 0.86, AA: 0.83, A: 0.79, BBB: 0.75, BB: 0.71, B: 0.69 } },\n };\n const LIM_MEDIDA = { VaR: 220, ES: null, Std: null }; // límite configurado SOLO para VaR (X-10)\n /* Overlays del scatter iso-risk de Marginal (F2.4): h = límite %-riesgo (marginal/VM),\n v = límite de tamaño de exposición (MM DOP), hyp = límite de riesgo absoluto POR POSICIÓN\n (MM DOP marginal; hipérbola X·Y=hyp — la recomendada). Solo VaR tiene h/hyp configurados:\n bajo ES/Std la leyenda muestra \"Límite: N/D\" y el KPI 3 es N/D aunque exista v (regla §10). */\n const MG_LIMITS = {\n VaR: { h: 4.5, v: 600, hyp: 8 },\n ES: { h: null, v: 600, hyp: null },\n Std: { h: null, v: 600, hyp: null },\n };\n const SUPUESTO_MARKERS = [{ fecha: \"03/01/2026\", label: \"Cambio de supuesto: ρ 20%→25% · S&P\" }];\n const RHO_PREV_FACTOR = 0.93; // cortes previos al marcador: ρ=20% → medida menor (discontinuidad visible)\n const SUPUESTO = { fuente: \"S&P\", metodo: \"ρ=25% axiomático\", tc: \"TC 61.50 (BCRD venta, 30/01/2026)\" };\n const AS_OF = \"30/01/2026\";\n const DATASET = \"calc-submit 1782766289\";\n /* F2.10 — Backtesting. Convención del mock (declarada, X-D1): cortes EOD diarios hábiles;\n VaR 99%·1a escalado al horizonte entre cortes por raíz de tiempo (÷√252) — la metodología\n oficial del tenant gobierna la comparación. Historia visible: 250 cortes bajo ρ=20%\n (sub-ventana homogénea CERRADA el 02/01/2026) + 20 cortes bajo ρ=25% (ventana válida\n vigente). El cambio 03/01/2026 (sábado, marcador compartido del mock) es EFECTIVO desde\n el corte hábil siguiente, 05/01/2026 (X-#2). VAR1A ≡ VaR total de Métricas de Riesgo\n (misma fórmula que mrAgg con medida VaR — constante única del mock, X-#9). */\n const BT = (() => {\n const VAR1A = positions.reduce((a, p) => {\n const vmDop = p.moneda === \"USD\" ? p.vm * TC : p.vm;\n return a + vmDop * RISK.VaR.rate[p.rating] * RISK.VaR.divF[p.rating];\n }, 0);\n const SQ = Math.sqrt(252);\n const fmt = d => String(d.getDate()).padStart(2, \"0\") + \"/\" + String(d.getMonth() + 1).padStart(2, \"0\") + \"/\" + d.getFullYear();\n let seed = 20260130;\n const rng = () => { seed = (seed * 1664525 + 1013904223) % 4294967296; return seed / 4294967296; };\n const days = [];\n const d = new Date(2026, 0, 30);\n while (days.length < 270) {\n if (d.getDay() !== 0 && d.getDay() !== 6) days.unshift(new Date(d));\n d.setDate(d.getDate() - 1);\n }\n const EFFECTIVE = \"05/01/2026\";\n const EXC = { \"14/03/2025\": 1.12, \"08/07/2025\": 1.05, \"09/10/2025\": 1.22, \"18/11/2025\": 1.08, \"26/01/2026\": 1.10 };\n let post = false;\n const cortes = days.map((dt, i) => {\n const f = fmt(dt);\n if (f === EFFECTIVE) post = true;\n const drift = 1 + 0.025 * Math.sin(i / 38) + (rng() - 0.5) * 0.012;\n const var1a = VAR1A * (post ? 1 : 0.93) * drift;\n const varEsc = var1a / SQ;\n let pnl = (rng() + rng() + rng() - 1.5) * varEsc * 0.72;\n if (EXC[f]) pnl = -varEsc * EXC[f];\n else if (pnl < -varEsc * 0.96) pnl = -varEsc * 0.9;\n return { f, pnl: +pnl.toFixed(2), varEsc: +varEsc.toFixed(2), valida: post };\n });\n return {\n cortes, EFFECTIVE, VAR1A: +VAR1A.toFixed(2), cambios: 1,\n cambioLabel: \"Cambio de supuesto: ρ 20%→25% · S&P (03/01/2026, efectivo corte 05/01/2026)\",\n prevWin: {\n rot: \"S&P · ρ=20% · 250 cortes (cerrada el 02/01/2026)\", exc: 4,\n kupiec: { lr: 0.755, p: 0.385, v: \"Aceptado\" },\n christo: { lr: 2.47, p: 0.29, v: \"Aceptado\" },\n basilea: { zona: \"Verde\", conteo: \"4/250\", mult: \"3.00\" },\n },\n mcPrecision: 3.2, // semiancho % del VaR @99% — cuadratura con el punto final del BL (X-D6)\n quantiles: { \"1%\": { q: 0.01, nivel: 1.00 }, \"0.5%\": { q: 0.005, nivel: 1.18 }, \"0.1%\": { q: 0.001, nivel: 1.45 } },\n };\n })();\n /* F2.8 — Factores de Riesgo. Atribución Euler por factor (pesos por rating, Σ=1 por fila:\n ratings bajos más default-driven, altos más migración/spread) ⇒ Σ factores×ratings = VaR\n total EXACTO (misma fórmula que mrAgg-VaR / BT.VAR1A — fuente única, X-#3).\n Prob ≥1 default COMPUTADA de las PD de TRANS por posición (coherencia por construcción, X-#9).\n Monotonía de ρ obligatoria (X-#8): RHO_ZERO_FACTOR 0.78 < RHO_PREV_FACTOR 0.93 < 1\n (el VaR crece con ρ: VaR(0) < VaR(20%) < VaR(25%)). */\n const FR = (() => {\n const W = {\n AAA: { mig: 0.52, def: 0.10, rec: 0.05, spr: 0.33 },\n AA: { mig: 0.50, def: 0.14, rec: 0.07, spr: 0.29 },\n A: { mig: 0.44, def: 0.24, rec: 0.10, spr: 0.22 },\n BBB: { mig: 0.36, def: 0.36, rec: 0.13, spr: 0.15 },\n BB: { mig: 0.28, def: 0.48, rec: 0.15, spr: 0.09 },\n B: { mig: 0.20, def: 0.58, rec: 0.16, spr: 0.06 },\n };\n const RHO_ZERO_FACTOR = 0.78; // VaR(ρ=0) = 0.78 · VaR(ρ=25%) ⇒ efecto correlación = 22% del VaR\n /* Matriz de transición mock a 1 año (%). Filas: rating de partida (D excluido — estado\n absorbente, sin migración de salida); columnas: llegada AAA…CCC, D. Cada fila Σ=100.\n PD (última columna) monotónica AAA→CCC: 0.01 → 22.73%. Única fuente de los umbrales Z\n de Merton (TR) y de la Prob ≥1 default (KPI 4). */\n const DEST = [\"AAA\", \"AA\", \"A\", \"BBB\", \"BB\", \"B\", \"CCC\", \"D\"];\n const TRANS = {\n AAA: [90.80, 8.30, 0.70, 0.10, 0.05, 0.03, 0.01, 0.01],\n AA: [0.70, 90.60, 7.80, 0.60, 0.15, 0.09, 0.04, 0.02],\n A: [0.09, 2.30, 91.00, 5.50, 0.70, 0.25, 0.10, 0.06],\n BBB: [0.02, 0.30, 5.90, 86.90, 5.30, 1.10, 0.30, 0.18],\n BB: [0.03, 0.14, 0.65, 7.70, 80.50, 8.80, 1.10, 1.08],\n B: [0.01, 0.11, 0.24, 0.45, 6.90, 83.50, 3.90, 4.89],\n CCC: [0.02, 0.05, 0.20, 0.30, 1.00, 11.20, 64.50, 22.73],\n };\n // Φ⁻¹ (aproximación de Acklam) — deriva los umbrales Z desde TRANS\n const invNorm = p => {\n const a = [-3.969683028665376e+01, 2.209460984245205e+02, -2.759285104469687e+02, 1.383577518672690e+02, -3.066479806614716e+01, 2.506628277459239e+00];\n const b = [-5.447609879822406e+01, 1.615858368580409e+02, -1.556989798598866e+02, 6.680131188771972e+01, -1.328068155288572e+01];\n const c = [-7.784894002430293e-03, -3.223964580411365e-01, -2.400758277161838e+00, -2.549732539343734e+00, 4.374664141464968e+00, 2.938163982698783e+00];\n const dd = [7.784695709041462e-03, 3.224671290700398e-01, 2.445134137142996e+00, 3.754408661907416e+00];\n const pl = 0.02425;\n if (p < pl) { const q = Math.sqrt(-2 * Math.log(p)); return (((((c[0] * q + c[1]) * q + c[2]) * q + c[3]) * q + c[4]) * q + c[5]) / ((((dd[0] * q + dd[1]) * q + dd[2]) * q + dd[3]) * q + 1); }\n if (p > 1 - pl) { const q = Math.sqrt(-2 * Math.log(1 - p)); return -(((((c[0] * q + c[1]) * q + c[2]) * q + c[3]) * q + c[4]) * q + c[5]) / ((((dd[0] * q + dd[1]) * q + dd[2]) * q + dd[3]) * q + 1); }\n const q = p - 0.5, r = q * q;\n return (((((a[0] * r + a[1]) * r + a[2]) * r + a[3]) * r + a[4]) * r + a[5]) * q / (((((b[0] * r + b[1]) * r + b[2]) * r + b[3]) * r + b[4]) * r + 1);\n };\n /* Z[r] = fronteras superiores de banda de llegada, de peor a mejor:\n [{dest:\"D\", zUp:Φ⁻¹(pD)}, {dest:\"CCC\", zUp:Φ⁻¹(pD+pCCC)}, …, {dest:\"AA\", zUp:Φ⁻¹(1−pAAA)}];\n la banda AAA es el resto superior. */\n const Z = {};\n Object.keys(TRANS).forEach(r => {\n const row = TRANS[r].map(x => x / 100);\n let acc = 0; const bnds = [];\n for (let i = row.length - 1; i >= 1; i--) { acc += row[i]; bnds.push({ dest: DEST[i], zUp: +invNorm(Math.min(acc, 1 - 1e-9)).toFixed(3) }); }\n Z[r] = bnds;\n });\n const PD = {}; Object.keys(TRANS).forEach(r => { PD[r] = TRANS[r][7] / 100; });\n // Prob de ≥1 default del libro a 1 año: 1 − Π(1 − PD_rating) sobre las posiciones\n const P_DEFAULT = 1 - positions.reduce((a, p) => a * (1 - PD[p.rating]), 1);\n return {\n W, RHO_ZERO_FACTOR, DEST, TRANS, Z, PD, P_DEFAULT,\n prevComparable: true, // cambio de supuesto efectivo 05/01/2026 < corte anterior 28/01 ⇒ Δ y chips comparables (rama N/D implementada, X-#6)\n RHO_UNIF: 25, // matriz uniforme ρ=25% axiomático (Alt 2): TODAS las celdas — el par intra-rating también es 25 (no es autocorrelación de un activo)\n };\n })();\n /* F2.9 — Sensibilidades. Barridos OAT precomputados del corte EOD con common random\n numbers (CRN), set de la metodología oficial (no editable). Factores multiplicativos\n sobre el VaR base (misma fórmula única mrAgg-VaR ≡ BT.VAR1A ≡ FR total — fuente única).\n Monotonía de ρ coherente con FR (X-#8 heredado): VaR(0)=0.78·b < VaR(15%)=0.885·b <\n VaR(20%)=0.93·b < b(25%) < VaR(35%)=1.13·b. wR = vector de reparto del ΔVaR por rating\n (Σ=1; ρ y σ cargan a ratings bajos, μ recovery carga donde vive la seniority) — el Δ de\n contribución Euler por rating es aditivo por construcción (BL/BR). Signo del barrido\n unilateral (fuente S&P→Moody's): +3.6% — la matriz Moody's del mock tiene PDs speculative-grade\n mayores (convención declarada del mock, X-12).\n prevComparable ÚNICO: se hereda de FR (mismo badge entre cortes). publishedUnderBadge\n gobierna la ancla badge↔barridos y DOMINA sobre prevComparable (X-1); en el mock es true\n y su rama degradada se demuestra con el estado \"partial\" del prototipo (X-7). */\n const SENS = (() => {\n const sweeps = [\n { id: \"rho\", lbl: \"ρ (correlación de activos)\", base: \"25%\", swp: \"15% / 35%\", fLo: 0.885, fHi: 1.130,\n wR: { AAA: 0.04, AA: 0.10, A: 0.24, BBB: 0.26, BB: 0.22, B: 0.14 } },\n { id: \"muSen\", lbl: \"μ recovery — senior\", base: \"45%\", swp: \"40% / 50%\", fLo: 1.072, fHi: 0.938,\n wR: { AAA: 0.08, AA: 0.16, A: 0.30, BBB: 0.24, BB: 0.14, B: 0.08 } },\n { id: \"sigSub\", lbl: \"σ recovery — subordinada\", base: \"28%\", swp: \"21% / 35%\", fLo: 0.962, fHi: 1.055,\n wR: { AAA: 0.02, AA: 0.06, A: 0.22, BBB: 0.30, BB: 0.26, B: 0.14 } },\n { id: \"muSub\", lbl: \"μ recovery — subordinada\", base: \"30%\", swp: \"25% / 35%\", fLo: 1.048, fHi: 0.960,\n wR: { AAA: 0.02, AA: 0.05, A: 0.20, BBB: 0.31, BB: 0.27, B: 0.15 } },\n { id: \"sigSen\", lbl: \"σ recovery — senior\", base: \"22%\", swp: \"16.5% / 27.5%\", fLo: 0.975, fHi: 1.038,\n wR: { AAA: 0.07, AA: 0.15, A: 0.29, BBB: 0.25, BB: 0.15, B: 0.09 } },\n { id: \"fuente\", lbl: \"Fuente de transición\", base: \"S&P\", swp: \"→ Moody's (alternativa única)\", fLo: null, fHi: 1.036, unilateral: true,\n wR: { AAA: 0.03, AA: 0.08, A: 0.22, BBB: 0.27, BB: 0.25, B: 0.15 } },\n ];\n // ΔVaR máx del corte anterior (factor sobre el VaR base anterior) — para la columna Var del BR\n const prevFmax = { rho: 0.126, muSen: 0.069, sigSub: 0.052, muSub: 0.046, sigSen: 0.036, fuente: 0.034 };\n return {\n sweeps, prevFmax,\n publishedUnderBadge: true, // ancla badge↔barridos (§13): true = el centro del tornado ES el supuesto del badge\n prevInfluyente: \"ρ (correlación de activos)\", // chip categórico del KPI 2 (comparable)\n };\n })();\n return { TC, TC_PREV, positions, RATINGS, TIPOS, DATES, HIST_TOTAL, HIST_SHARE, LIMITES, RISK, LIM_MEDIDA, MG_LIMITS, SUPUESTO_MARKERS, RHO_PREV_FACTOR, SUPUESTO, AS_OF, DATASET, BT, FR, SENS };\n})();\n",[159],"e797f993-49ef-4cb4-8219-a6ced7c60297",["D",1787332301862],50,[28],"pulls","branches","forks","envVars",{"_177":63,"_179":63},"connections",{"_177":63,"_178":63},"count","unfilledCount","emptyCount",{"_43":166,"_32":28,"_30":31,"_135":-5,"_41":42,"_40":14,"_137":25,"_38":198,"_13":168,"_33":34,"_35":37,"_24":25,"_141":-5,"_186":199,"_27":200},{"_43":158,"_32":159,"_30":160,"_135":-5,"_41":42,"_40":161,"_137":25,"_38":195,"_13":163,"_33":34,"_35":164,"_24":25,"_141":-5,"_186":196,"_27":197},{"_43":150,"_32":151,"_30":152,"_135":-5,"_41":42,"_40":153,"_137":59,"_38":192,"_13":155,"_33":20,"_35":156,"_24":25,"_141":-5,"_186":193,"_27":194},{"_43":143,"_32":144,"_30":57,"_135":-5,"_41":42,"_40":145,"_137":59,"_38":189,"_13":63,"_33":147,"_35":148,"_24":25,"_141":-5,"_186":190,"_27":191},{"_43":132,"_32":133,"_30":134,"_135":-5,"_41":42,"_40":136,"_137":25,"_38":185,"_13":139,"_33":20,"_35":140,"_24":25,"_141":-5,"_186":187,"_27":188},["D",1787331497474],"children",[],[133],["D",1787326725014],[],[144],["D",1787327075110],[],[151],["D",1787332294375],[],[159],["D",1787332301862],[],[28],["D",1787326658800],{"_97":96,"_104":100},{"_97":96,"_98":-5,"_99":100,"_18":55,"_101":102,"_103":25},{"_13":86,"_32":87,"_43":42,"_88":44,"_89":-5,"_90":-5,"_91":-5,"_92":-5,"_93":-5,"_94":45,"_95":96},"ctx",{"_79":-5,"_124":238,"_239":240,"_241":242,"_243":25,"_244":25,"_245":25,"_246":59,"_247":-5,"_248":-7,"_249":-7,"_250":63,"_251":25,"_252":25,"_253":-5,"_254":25,"_255":25,"_256":25,"_257":-5},"ENV",{"_209":210,"_211":212,"_213":87,"_214":215,"_216":217,"_218":219,"_220":221,"_222":223,"_224":225,"_226":227,"_228":229,"_230":231,"_232":233,"_234":235,"_236":237},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

Diff · v70