Files
worldmonitor/scripts/seed-fear-greed.mjs
Elie Habib 7013b2f9f1 feat(market): Fear & Greed Index 2.0 — 10-category composite sentiment panel (#2181)
* Add Fear & Greed Index 2.0 reverse engineering brief

Analyzes the 10-category weighted composite (Sentiment, Volatility,
Positioning, Trend, Breadth, Momentum, Liquidity, Credit, Macro,
Cross-Asset) with scoring formulas, data source audit, and
implementation plan for building it as a worldmonitor panel.

https://claude.ai/code/session_01HR69u6oF1VCMwsC2PHFL8i

* Add seed script implementation plan to F&G brief

Details exact endpoints, Yahoo symbols (17 calls), Redis key schema,
computed metrics, FRED series to add (BAMLC0A0CM, SOFR), CNN/AAII
sources, output JSON schema, and estimated runtime (~8s per seed run).

https://claude.ai/code/session_01HR69u6oF1VCMwsC2PHFL8i

* Update brief: all sources are free, zero paid APIs needed

- CBOE CDN CSVs for put/call ratios (totalpc.csv, equitypc.csv)
- CNN dataviz API for Fear & Greed (production.dataviz.cnn.io)
- Yahoo Finance for VIX9D/VIX3M/SKEW/RSP/NYA (standard symbols)
- FRED for IG spread (BAMLC0A0CM) and SOFR (add to existing array)
- AAII scrape for bull/bear survey (only medium-effort source)
- Breadth via RSP/SPY divergence + NYSE composite (no scraping)

https://claude.ai/code/session_01HR69u6oF1VCMwsC2PHFL8i

* Add verified Yahoo symbols for breadth + finalized source list

New discoveries:
- ^MMTH = % stocks above 200 DMA (direct Yahoo symbol!)
- C:ISSU = NYSE advance/decline data
- CNN endpoint accepts date param for historical data
- CBOE CSVs have data back to 2003
- 33 total calls per seed run, ~6s runtime

All 10 categories now have confirmed free sources.

https://claude.ai/code/session_01HR69u6oF1VCMwsC2PHFL8i

* Rewrite F&G brief as forward-looking design doc

Remove all reverse-engineering language, screenshot references, and
discovery notes. Clean structure: goal, scoring model, data sources,
formulas, seed script plan, implementation phases, MVP path.

https://claude.ai/code/session_01HR69u6oF1VCMwsC2PHFL8i

* docs: apply gold standard corrections to fear-greed-index-2.0 brief

* feat(market): add Fear & Greed Index 2.0 — 10-category composite sentiment panel

Composite 0-100 index from 10 weighted categories: sentiment (CNN F&G,
AAII, crypto F&G), volatility (VIX, term structure), positioning (P/C
ratio, SKEW), trend (SPX vs MAs), breadth (% >200d, RSP/SPY divergence),
momentum (sector RSI, ROC), liquidity (M2, Fed BS, SOFR), credit (HY/IG
spreads), macro (Fed rate, yield curve, unemployment), cross-asset
(gold/bonds/DXY vs equities).

Data layer:
- seed-fear-greed.mjs: 19 Yahoo symbols (150ms gaps), CBOE P/C CSVs,
  CNN F&G API, AAII scrape (degraded-safe), FRED Redis reads. TTL 64800s.
- seed-economy.mjs: add BAMLC0A0CM (IG spread) and SOFR to FRED_SERIES.
- Bootstrap 4-file checklist: cache-keys, bootstrap.js, health.js, handler.

Proto + RPC:
- get_fear_greed_index.proto with FearGreedCategory message.
- get-fear-greed-index.ts handler reads seeded Redis data.

Frontend:
- FearGreedPanel with gauge, 9-metric header grid, 10-category breakdown.
- Self-loading via bootstrap hydration + RPC fallback.
- Registered in panel-layout, App.ts (prime + refresh), panel config,
  Cmd-K commands, finance variant, i18n (en/ar/zh/es).

* fix(market): add RPC_CACHE_TIER entry for get-fear-greed-index

* fix(docs): escape bare angle bracket in fear-greed brief for MDX

* fix(docs): fix markdown lint errors in fear-greed brief (blank lines around headings/lists)

* fix(market): fix seed-fear-greed bugs from code review

- fredLatest/fredNMonthsAgo: guard parseFloat with Number.isFinite to
  handle FRED's "." missing-data sentinel (was returning NaN which
  propagated through scoring as a truthy non-null value)
- Remove 3 unused Yahoo symbols (^NYA, HYG, LQD) that were fetched
  but not referenced in any scoring category (saves ~450ms per run)
- fedRateStr: display effective rate directly instead of deriving
  target range via (fedRate - 0.25) which was incorrect

* fix(market): address P2/P3 review findings in Fear & Greed

- FearGreedPanel: add mapSeedPayload() to correctly map raw seed
  JSON to proto-shaped FearGreedData; bootstrap hydration was always
  falling through to RPC because seed shape (composite.score) differs
  from proto shape (compositeScore)
- FearGreedPanel: fix fmt() — remove === 0 guard and add explicit
  > 0 checks on VIX and P/C Ratio display to handle proto default
  zeros without masking genuine zero values (e.g. pctAbove200d)
- seed-fear-greed: remove broken history write — each run overwrote
  the key with a single-entry array (no read-then-append), making the
  90-day TTL meaningless; no consumer exists yet so defer to later
- seed-fear-greed: extract hySpreadVal const to avoid double fredLatest call
- seed-fear-greed: fix stale comment (19 symbols → 16 after prior cleanup)

---------

Co-authored-by: Claude <noreply@anthropic.com>
2026-03-24 09:45:59 +04:00

412 lines
20 KiB
JavaScript

#!/usr/bin/env node
import { loadEnvFile, CHROME_UA, runSeed, readSeedSnapshot, sleep } from './_seed-utils.mjs';
loadEnvFile(import.meta.url);
const FEAR_GREED_KEY = 'market:fear-greed:v1';
const FEAR_GREED_TTL = 64800; // 18h = 3x 6h interval
const FRED_PREFIX = 'economic:fred:v1';
// --- Yahoo Finance fetching (16 symbols, 150ms gaps) ---
const YAHOO_SYMBOLS = ['^GSPC','^VIX','^VIX9D','^VIX3M','^SKEW','^MMTH','C:ISSU','GLD','TLT','SPY','RSP','DX-Y.NYB','XLK','XLF','XLE','XLV'];
async function fetchYahooSymbol(symbol) {
const url = `https://query1.finance.yahoo.com/v8/finance/chart/${encodeURIComponent(symbol)}?interval=1d&range=3mo`;
try {
const resp = await fetch(url, {
headers: { 'User-Agent': CHROME_UA, Accept: 'application/json' },
signal: AbortSignal.timeout(10_000),
});
if (!resp.ok) { console.warn(` Yahoo ${symbol}: HTTP ${resp.status}`); return null; }
const data = await resp.json();
const result = data?.chart?.result?.[0];
if (!result) return null;
const closes = result.indicators?.quote?.[0]?.close ?? [];
const validCloses = closes.filter(v => v != null);
const price = result.meta?.regularMarketPrice ?? validCloses.at(-1) ?? null;
return { symbol, price, closes: validCloses };
} catch (e) {
console.warn(` Yahoo ${symbol}: ${e.message}`);
return null;
}
}
async function fetchAllYahoo() {
const results = {};
for (const sym of YAHOO_SYMBOLS) {
results[sym] = await fetchYahooSymbol(sym);
await sleep(150);
}
return results;
}
// --- CBOE P/C ratios ---
async function fetchCBOE() {
const [totalResp, equityResp] = await Promise.allSettled([
fetch('https://cdn.cboe.com/api/global/us_indices/daily_prices/totalpc.csv', { headers: { 'User-Agent': CHROME_UA }, signal: AbortSignal.timeout(10_000) }),
fetch('https://cdn.cboe.com/api/global/us_indices/daily_prices/equitypc.csv', { headers: { 'User-Agent': CHROME_UA }, signal: AbortSignal.timeout(10_000) }),
]);
const parseLastValue = async (resp) => {
if (resp.status !== 'fulfilled' || !resp.value.ok) return null;
const text = await resp.value.text();
const lines = text.trim().split('\n').filter(l => l.trim());
const last = lines.at(-1)?.split(',');
return last?.length >= 2 ? parseFloat(last[1]) : null;
};
const [totalPc, equityPc] = await Promise.all([parseLastValue(totalResp), parseLastValue(equityResp)]);
return { totalPc, equityPc };
}
// --- CNN Fear & Greed ---
async function fetchCNN() {
try {
const date = new Date().toISOString().slice(0,10).replace(/-/g,'');
const resp = await fetch(`https://production.dataviz.cnn.io/index/fearandgreed/graphdata/${date}`, {
headers: { 'User-Agent': CHROME_UA, Accept: 'application/json' },
signal: AbortSignal.timeout(8_000),
});
if (!resp.ok) return null;
const data = await resp.json();
const score = data?.fear_and_greed?.score;
const rating = data?.fear_and_greed?.rating;
return score != null ? { score: Math.round(score), label: rating ?? labelFromScore(Math.round(score)) } : null;
} catch { return null; }
}
// --- AAII Sentiment (LOW reliability, always wrapped, non-blocking) ---
async function fetchAAII() {
try {
const resp = await fetch('https://www.aaii.com/sentimentsurvey/sent_results', {
headers: { 'User-Agent': CHROME_UA, Accept: 'text/html,application/xhtml+xml' },
signal: AbortSignal.timeout(8_000),
});
if (!resp.ok) return null;
const html = await resp.text();
const bullMatch = html.match(/Bullish[^%]*?([\d.]+)%/i);
const bearMatch = html.match(/Bearish[^%]*?([\d.]+)%/i);
if (!bullMatch || !bearMatch) return null;
return { bull: parseFloat(bullMatch[1]), bear: parseFloat(bearMatch[1]) };
} catch (e) {
console.warn(' AAII: fetch failed:', e.message, '(using degraded Sentiment)');
return null;
}
}
// --- FRED Redis reads ---
async function readFred(seriesId) {
const url = process.env.UPSTASH_REDIS_REST_URL;
const token = process.env.UPSTASH_REDIS_REST_TOKEN;
if (!url || !token) return null;
try {
const resp = await fetch(`${url}/get/${encodeURIComponent(`${FRED_PREFIX}:${seriesId}:0`)}`, {
headers: { Authorization: `Bearer ${token}` },
signal: AbortSignal.timeout(5_000),
});
if (!resp.ok) return null;
const { result } = await resp.json();
if (!result) return null;
const parsed = JSON.parse(result);
const obs = parsed?.series?.observations;
if (!obs?.length) return null;
return obs;
} catch { return null; }
}
async function readMacroSignals() {
const url = process.env.UPSTASH_REDIS_REST_URL;
const token = process.env.UPSTASH_REDIS_REST_TOKEN;
if (!url || !token) return null;
try {
const resp = await fetch(`${url}/get/${encodeURIComponent('economic:macro-signals:v1')}`, {
headers: { Authorization: `Bearer ${token}` },
signal: AbortSignal.timeout(5_000),
});
if (!resp.ok) return null;
const { result } = await resp.json();
return result ? JSON.parse(result) : null;
} catch { return null; }
}
// --- Math helpers ---
function clamp(v, lo, hi) { return Math.max(lo, Math.min(hi, v)); }
function sma(prices, period) {
if (prices.length < period) return null;
return prices.slice(-period).reduce((a,b) => a+b, 0) / period;
}
function roc(prices, period) {
if (prices.length < period+1) return null;
const prev = prices[prices.length - period - 1];
const curr = prices[prices.length - 1];
return prev ? ((curr - prev) / prev) * 100 : null;
}
function rsi(prices, period=14) {
if (prices.length < period+1) return 50;
let gains=0, losses=0;
for (let i=prices.length-period; i<prices.length; i++) {
const d = prices[i]-prices[i-1];
if (d>0) gains+=d; else losses+=Math.abs(d);
}
if (losses===0) return 100;
const rs = (gains/period)/(losses/period);
return 100 - (100/(1+rs));
}
function fredLatest(obs) {
if (!obs) return null;
const v = parseFloat(obs.at(-1)?.value ?? 'NaN');
return Number.isFinite(v) ? v : null;
}
function fredNMonthsAgo(obs, months) {
if (!obs) return null;
const idx = obs.length - 1 - months;
if (idx < 0) return null;
const v = parseFloat(obs[idx]?.value ?? 'NaN');
return Number.isFinite(v) ? v : null;
}
function labelFromScore(s) {
if (s <= 20) return 'Extreme Fear';
if (s <= 40) return 'Fear';
if (s <= 60) return 'Neutral';
if (s <= 80) return 'Greed';
return 'Extreme Greed';
}
// --- Scoring ---
function scoreCategory(name, inputs) {
switch(name) {
case 'sentiment': {
const { cnnFg, aaiBull, aaiBear, cryptoFg } = inputs;
const degraded = aaiBull == null || aaiBear == null;
let score;
if (!degraded) {
const bullPct = clamp(aaiBull, 0, 100);
const bearPct = clamp(aaiBear, 0, 100);
const bullPercentile = clamp((bullPct / 60) * 100, 0, 100);
const bearPercentile = clamp((bearPct / 55) * 100, 0, 100);
if (cnnFg != null) {
score = (cnnFg * 0.4) + (bullPercentile * 0.3) + ((100 - bearPercentile) * 0.3);
} else {
score = (bullPercentile * 0.5) + ((100 - bearPercentile) * 0.5);
}
} else if (cnnFg != null) {
score = cnnFg;
} else if (cryptoFg != null) {
score = cryptoFg;
} else {
score = 50;
}
return { score: clamp(Math.round(score), 0, 100), degraded, inputs: { cnnFearGreed: cnnFg, aaiBull: aaiBull ?? null, aaiBear: aaiBear ?? null, cryptoFg } };
}
case 'volatility': {
const { vix, vix9d, vix3m } = inputs;
if (vix == null) return { score: 50, inputs };
const vixScore = clamp(100 - ((vix - 12) / 28) * 100, 0, 100);
const termScore = (vix9d != null && vix3m != null) ? (vix / vix3m < 1 ? 70 : 30) : 50;
const termStructure = (vix9d != null && vix3m != null) ? (vix / vix3m < 1 ? 'contango' : 'backwardation') : 'unknown';
return { score: Math.round(vixScore * 0.7 + termScore * 0.3), inputs: { vix, vix9d, vix3m, termStructure } };
}
case 'positioning': {
const { totalPc, equityPc, skew } = inputs;
const pc = totalPc ?? equityPc;
if (pc == null && skew == null) return { score: 50, inputs };
const pcScore = pc != null ? clamp(100 - ((pc - 0.7) / 0.6) * 100, 0, 100) : 50;
const skewScore = skew != null ? clamp(100 - ((skew - 100) / 50) * 100, 0, 100) : 50;
const w = pc != null && skew != null ? [0.6, 0.4] : [1.0, 0.0];
return { score: Math.round(pcScore * w[0] + skewScore * w[1]), inputs: { putCallRatio: pc, skew } };
}
case 'trend': {
const { prices } = inputs;
if (!prices?.length) return { score: 50, inputs: {} };
const price = prices.at(-1);
const s20 = sma(prices, 20), s50 = sma(prices, 50), s200 = sma(prices, 200);
const aboveCount = [s20, s50, s200].filter(s => s != null && price > s).length;
const dist200 = s200 ? (price - s200) / s200 : 0;
const score = (aboveCount / 3) * 50 + clamp(dist200 * 500 + 50, 0, 100) * 0.5;
return { score: Math.round(clamp(score, 0, 100)), inputs: { spxPrice: price, sma20: s20, sma50: s50, sma200: s200, aboveMaCount: aboveCount } };
}
case 'breadth': {
const { mmthPrice, rspCloses, spyCloses, advDecRatio } = inputs;
const breadthScore = mmthPrice != null ? clamp(mmthPrice, 0, 100) : 50;
const rspRoc = (rspCloses?.length && spyCloses?.length) ? (roc(rspCloses, 30) ?? 0) - (roc(spyCloses, 30) ?? 0) : null;
const rspScore = rspRoc != null ? clamp(rspRoc * 10 + 50, 0, 100) : 50;
const adScore = advDecRatio != null ? clamp((advDecRatio - 0.5) / 1.5 * 100, 0, 100) : 50;
const hasAd = advDecRatio != null;
const w = hasAd ? [0.4, 0.3, 0.3] : [0.57, 0, 0.43];
const score = breadthScore * w[0] + adScore * w[1] + rspScore * w[2];
return { score: Math.round(clamp(score, 0, 100)), inputs: { pctAbove200d: mmthPrice, rspSpyRatio: rspRoc, advDecRatio: advDecRatio ?? null } };
}
case 'momentum': {
const { spxCloses, sectorCloses } = inputs;
const spxRoc = spxCloses?.length ? roc(spxCloses, 20) : null;
const rocScore = spxRoc != null ? clamp(spxRoc * 10 + 50, 0, 100) : 50;
const sectorRsiValues = sectorCloses ? Object.values(sectorCloses).filter(Boolean).map(c => rsi(c)) : [];
const avgRsi = sectorRsiValues.length ? sectorRsiValues.reduce((a,b)=>a+b,0)/sectorRsiValues.length : 50;
const rsiScore = clamp((avgRsi - 30) / 40 * 100, 0, 100);
return { score: Math.round((rsiScore * 0.5 + rocScore * 0.5)), inputs: { spxRoc20d: spxRoc, sectorRsiAvg: Math.round(avgRsi) } };
}
case 'liquidity': {
const { m2Obs, walclObs, sofr } = inputs;
const m2Latest = fredLatest(m2Obs), m2Ago = fredNMonthsAgo(m2Obs, 12);
const m2Yoy = (m2Latest && m2Ago && m2Ago !== 0) ? ((m2Latest - m2Ago) / m2Ago) * 100 : null;
const walclLatest = fredLatest(walclObs), walclAgo = fredNMonthsAgo(walclObs, 1);
const fedBsMom = (walclLatest && walclAgo && walclAgo !== 0) ? ((walclLatest - walclAgo) / walclAgo) * 100 : null;
const m2Score = m2Yoy != null ? clamp(m2Yoy * 10 + 50, 0, 100) : 50;
const fedScore = fedBsMom != null ? clamp(fedBsMom * 20 + 50, 0, 100) : 50;
const sofrScore = sofr != null ? clamp(100 - sofr * 15, 0, 100) : 50;
return { score: Math.round(m2Score * 0.4 + fedScore * 0.3 + sofrScore * 0.3), inputs: { m2Yoy, fedBsMom, sofr } };
}
case 'credit': {
const { hyObs, igObs } = inputs;
const hySpread = fredLatest(hyObs), igSpread = fredLatest(igObs);
const hyScore = hySpread != null ? clamp(100 - ((hySpread - 3.0) / 5.0) * 100, 0, 100) : 50;
const igScore = igSpread != null ? clamp(100 - ((igSpread - 0.8) / 2.0) * 100, 0, 100) : 50;
const hyPrev = fredNMonthsAgo(hyObs, 1);
const hyTrend = (hySpread != null && hyPrev != null) ? (hySpread < hyPrev ? 'narrowing' : hySpread > hyPrev ? 'widening' : 'stable') : 'stable';
const trendScore = hyTrend === 'narrowing' ? 70 : hyTrend === 'widening' ? 30 : 50;
return { score: Math.round(hyScore * 0.4 + igScore * 0.3 + trendScore * 0.3), inputs: { hySpread, igSpread, hyTrend30d: hyTrend } };
}
case 'macro': {
const { fedObs, curveObs, unrateObs } = inputs;
const fedRate = fredLatest(fedObs), t10y2y = fredLatest(curveObs), unrate = fredLatest(unrateObs);
const rateScore = fedRate != null ? clamp(100 - fedRate * 15, 0, 100) : 50;
const curveScore = t10y2y != null ? (t10y2y > 0 ? clamp(60 + t10y2y * 20, 0, 100) : clamp(40 + t10y2y * 40, 0, 100)) : 50;
const unempScore = unrate != null ? clamp(100 - (unrate - 3.5) * 20, 0, 100) : 50;
return { score: Math.round(rateScore * 0.3 + curveScore * 0.4 + unempScore * 0.3), inputs: { fedRate, t10y2y, unrate } };
}
case 'crossAsset': {
const { gldCloses, tltCloses, spyCloses, dxyCloses } = inputs;
const goldRoc = gldCloses?.length ? roc(gldCloses, 30) : null;
const tltRoc = tltCloses?.length ? roc(tltCloses, 30) : null;
const spyRoc = spyCloses?.length ? roc(spyCloses, 30) : null;
const dxyRoc = dxyCloses?.length ? roc(dxyCloses, 30) : null;
const goldSignal = (goldRoc != null && spyRoc != null) ? (goldRoc > spyRoc ? 30 : 70) : 50;
const bondSignal = (tltRoc != null && spyRoc != null) ? (tltRoc > spyRoc ? 30 : 70) : 50;
const dxySignal = dxyRoc != null ? (dxyRoc > 0 ? 40 : 60) : 50;
return { score: Math.round((goldSignal + bondSignal + dxySignal) / 3), inputs: { goldReturn30d: goldRoc, tltReturn30d: tltRoc, spyReturn30d: spyRoc, dxyChange30d: dxyRoc } };
}
default: return { score: 50, inputs };
}
}
const WEIGHTS = { sentiment: 0.10, volatility: 0.10, positioning: 0.15, trend: 0.10, breadth: 0.10, momentum: 0.10, liquidity: 0.15, credit: 0.10, macro: 0.05, crossAsset: 0.05 };
async function fetchAll() {
const prevSnapshot = await readSeedSnapshot(FEAR_GREED_KEY).catch(() => null);
const previousScore = prevSnapshot?.composite?.score ?? null;
const [yahooResults, cboeResult, cnnResult, aaiiResult, macroSignals] = await Promise.allSettled([
fetchAllYahoo(),
fetchCBOE(),
fetchCNN(),
fetchAAII(),
readMacroSignals(),
]);
const yahoo = yahooResults.status === 'fulfilled' ? yahooResults.value : {};
const cboe = cboeResult.status === 'fulfilled' ? cboeResult.value : {};
const cnn = cnnResult.status === 'fulfilled' ? cnnResult.value : null;
const aaii = aaiiResult.status === 'fulfilled' ? aaiiResult.value : null;
const macro = macroSignals.status === 'fulfilled' ? macroSignals.value : null;
if (yahooResults.status === 'rejected') console.warn(' Yahoo batch failed:', yahooResults.reason?.message);
if (cboeResult.status === 'rejected') console.warn(' CBOE failed:', cboeResult.reason?.message);
if (cnnResult.status === 'rejected') console.warn(' CNN failed:', cnnResult.reason?.message);
if (aaiiResult.status === 'rejected') console.warn(' AAII failed:', aaiiResult.reason?.message);
const [hyObs, igObs, m2Obs, walclObs, sofrObs, fedObs, curveObs, unrateObs, vixObs, dgs10Obs] = await Promise.all([
readFred('BAMLH0A0HYM2'), readFred('BAMLC0A0CM'), readFred('M2SL'), readFred('WALCL'),
readFred('SOFR'), readFred('FEDFUNDS'), readFred('T10Y2Y'), readFred('UNRATE'), readFred('VIXCLS'), readFred('DGS10'),
]);
const gspc = yahoo['^GSPC'];
const vixData = yahoo['^VIX'];
const vix9d = yahoo['^VIX9D'];
const vix3m = yahoo['^VIX3M'];
const skew = yahoo['^SKEW'];
const mmth = yahoo['^MMTH'];
const cissu = yahoo['C:ISSU'];
const gld = yahoo['GLD'], tlt = yahoo['TLT'], spy = yahoo['SPY'], rsp = yahoo['RSP'];
const dxy = yahoo['DX-Y.NYB'];
const xlk = yahoo['XLK'], xlf = yahoo['XLF'], xle = yahoo['XLE'], xlv = yahoo['XLV'];
const vixLive = vixData?.price ?? fredLatest(vixObs);
const vix9dPrice = vix9d?.price ?? null;
const vix3mPrice = vix3m?.price ?? null;
const skewPrice = skew?.price ?? null;
const mmthPrice = mmth?.price ?? null;
const sofrRate = fredLatest(sofrObs);
const cryptoFg = macro?.fearGreed?.score ?? macro?.signals?.fearGreed?.value ?? null;
let advDecRatio = null;
if (cissu?.price != null) {
advDecRatio = cissu.price > 0 ? Math.min(cissu.price / 100, 2.0) : null;
}
const cats = {
sentiment: scoreCategory('sentiment', { cnnFg: cnn?.score ?? null, aaiBull: aaii?.bull ?? null, aaiBear: aaii?.bear ?? null, cryptoFg }),
volatility: scoreCategory('volatility', { vix: vixLive, vix9d: vix9dPrice, vix3m: vix3mPrice }),
positioning: scoreCategory('positioning', { totalPc: cboe.totalPc, equityPc: cboe.equityPc, skew: skewPrice }),
trend: scoreCategory('trend', { prices: gspc?.closes ?? [] }),
breadth: scoreCategory('breadth', { mmthPrice, rspCloses: rsp?.closes, spyCloses: spy?.closes, advDecRatio }),
momentum: scoreCategory('momentum', { spxCloses: gspc?.closes, sectorCloses: { XLK: xlk?.closes, XLF: xlf?.closes, XLE: xle?.closes, XLV: xlv?.closes } }),
liquidity: scoreCategory('liquidity', { m2Obs, walclObs, sofr: sofrRate }),
credit: scoreCategory('credit', { hyObs, igObs }),
macro: scoreCategory('macro', { fedObs, curveObs, unrateObs }),
crossAsset: scoreCategory('crossAsset', { gldCloses: gld?.closes, tltCloses: tlt?.closes, spyCloses: spy?.closes, dxyCloses: dxy?.closes }),
};
const compositeScore = Math.round(
Object.entries(cats).reduce((sum, [name, cat]) => sum + cat.score * WEIGHTS[name], 0) * 10
) / 10;
const compositeLabel = labelFromScore(compositeScore);
const fedRate = fredLatest(fedObs);
const fedRateStr = fedRate != null ? `${fedRate.toFixed(2)}%` : null;
const hySpreadVal = fredLatest(hyObs);
const payload = {
timestamp: new Date().toISOString(),
composite: { score: compositeScore, label: compositeLabel, previous: previousScore },
categories: {
sentiment: { score: cats.sentiment.score, weight: WEIGHTS.sentiment, contribution: Math.round(cats.sentiment.score * WEIGHTS.sentiment * 10)/10, inputs: cats.sentiment.inputs, degraded: cats.sentiment.degraded ?? false },
volatility: { score: cats.volatility.score, weight: WEIGHTS.volatility, contribution: Math.round(cats.volatility.score * WEIGHTS.volatility * 10)/10, inputs: cats.volatility.inputs },
positioning: { score: cats.positioning.score, weight: WEIGHTS.positioning, contribution: Math.round(cats.positioning.score * WEIGHTS.positioning * 10)/10, inputs: cats.positioning.inputs },
trend: { score: cats.trend.score, weight: WEIGHTS.trend, contribution: Math.round(cats.trend.score * WEIGHTS.trend * 10)/10, inputs: cats.trend.inputs },
breadth: { score: cats.breadth.score, weight: WEIGHTS.breadth, contribution: Math.round(cats.breadth.score * WEIGHTS.breadth * 10)/10, inputs: cats.breadth.inputs },
momentum: { score: cats.momentum.score, weight: WEIGHTS.momentum, contribution: Math.round(cats.momentum.score * WEIGHTS.momentum * 10)/10, inputs: cats.momentum.inputs },
liquidity: { score: cats.liquidity.score, weight: WEIGHTS.liquidity, contribution: Math.round(cats.liquidity.score * WEIGHTS.liquidity * 10)/10, inputs: cats.liquidity.inputs },
credit: { score: cats.credit.score, weight: WEIGHTS.credit, contribution: Math.round(cats.credit.score * WEIGHTS.credit * 10)/10, inputs: cats.credit.inputs },
macro: { score: cats.macro.score, weight: WEIGHTS.macro, contribution: Math.round(cats.macro.score * WEIGHTS.macro * 10)/10, inputs: cats.macro.inputs },
crossAsset: { score: cats.crossAsset.score, weight: WEIGHTS.crossAsset, contribution: Math.round(cats.crossAsset.score * WEIGHTS.crossAsset * 10)/10, inputs: cats.crossAsset.inputs },
},
headerMetrics: {
cnnFearGreed: cnn ? { value: cnn.score, label: cnn.label } : null,
aaiBear: aaii ? { value: Math.round(aaii.bear), context: `${aaii.bear.toFixed(1)}%` } : null,
aaiBull: aaii ? { value: Math.round(aaii.bull), context: `${aaii.bull.toFixed(1)}%` } : null,
putCall: cboe.totalPc != null ? { value: cboe.totalPc } : null,
vix: vixLive != null ? { value: vixLive } : null,
hySpread: hySpreadVal != null ? { value: hySpreadVal } : null,
pctAbove200d: mmthPrice != null ? { value: mmthPrice } : null,
yield10y: fredLatest(dgs10Obs) != null ? { value: fredLatest(dgs10Obs) } : null,
fedRate: fedRateStr ? { value: fedRateStr } : null,
},
unavailable: false,
};
return payload;
}
function validate(data) {
return data?.composite?.score != null && data.timestamp != null;
}
runSeed('market', 'fear-greed', FEAR_GREED_KEY, fetchAll, {
validateFn: validate,
ttlSeconds: FEAR_GREED_TTL,
sourceVersion: 'yahoo-cboe-cnn-fred-v1',
}).catch((err) => {
console.error('FATAL:', err.message || err);
process.exit(1);
});