Tengu
Sign up
OverviewProductsPrivate MarketsTengu BrainAPI Reference

Risk Analytics

Institutional-grade intraday Value-at-Risk, stress scenarios, hedging signals, and pairs analytics for portfolio risk management.

What's inside#

Intraday VaR/CVaR on the live top-decile shadow book, computed via consensus ensemble (Cornish-Fisher + t-copula Monte Carlo + filtered historical simulation) updated sub-minute. Stress replays of 5 canonical crash scenarios (COVID, GFC, rate shock, flash crash, stagflation) on the live book with portfolio drawdown estimates. Institutional hedging signals computed from dealer positioning (gamma net, IV rank, 25-delta risk reversals) for index protection and collars. Pairs analytics (correlation, Kalman-filtered z-scores) for stat-arb setup detection. Single-name risk (ATR-based position sizing, regime-adjusted Kelly fractions) for pre-trade tickets.

Access#

5 credits/call, Pro and up.

Endpoints#

MethodPathDescription
GET/api/v2/risk/varPortfolio Value-at-Risk + CVaR (consensus Cornish-Fisher + t-copula + FHS). Horizon 1–30 days, confidence 50–99.99%.
GET/api/v3/risk/portfolio_varPortfolio VaR + CVaR — first-class lean tool (1-day horizon, liquidity-adjusted).
GET/api/v2/risk/stressStress-test scenarios (COVID, GFC, rate shock, flash crash, stagflation) replayed on live book.
GET/api/v3/institutional/hedgingIndex hedge proposals (SPY/QQQ/IWM) — protective puts, collars — from dealer gamma & IV rank.
GET/api/v2/pairs/kalmanKalman-filtered pairs signals (z-score, hedge ratio). 36-month correlation candidates fallback.
GET/api/risk/{ticker}Single-name pre-trade risk: ATR position size, regime-adjusted Kelly fraction, prediction checklist.

Examples#

Bash — get portfolio VaR at 99% confidence:

Shell
1curl -H "Authorization: Bearer $TENGU_API_KEY" \2  "https://firm.tengu.co/api/v2/risk/var?confidence=0.99&horizon_days=1"

Response shape:

JSON
1{2  "ok": true,3  "timestamp": "2026-07-05T14:22:33.456789+00:00",4  "as_of_ts": "2026-07-05T14:22:00+00:00",5  "horizon_days": 1,6  "confidence": 0.99,7  "var": 12500.50,8  "cvar": 18750.75,9  "method": "consensus(Cornish-Fisher, t-copula MC, filtered historical simulation)",10  "detail": {11    "cornish_fisher_var_99": 12200.00,12    "t_copula_var_99": 12800.30,13    "fhs_var_99": 12400.20,14    "liquidity_adjusted_var_99": 13100.00,15    "liquidity_haircut_pct": 8.0,16    "portfolio_value_usd": 250000.0017  },18  "portfolio": "top-decile equal-weight shadow book",19  "source": "gold.realtime_risk"20}

Python — get institutional hedging proposals:

Python
1import os, requests2 3r = requests.get(4    "https://firm.tengu.co/api/v3/institutional/hedging",5    headers={"Authorization": f"Bearer {os.environ['TENGU_API_KEY']}"},6    timeout=30,7)8r.raise_for_status()9data = r.json()10# data: {"ok": true, "timestamp": ..., "kind": "index_hedge_proposals", "hedge_set": ["SPY", "QQQ", "IWM"], "proposals": [...], "market_context": {...}, "source": "computed:options_signals+uw_dealer_greeks"}11proposals = data.get("proposals", [])12for p in proposals:13    print(f"{p['instrument']} — {p['hedge_type']}: {p['action']}")

Related#