PRE-MARKET · plan the next session
Use the latest completed session’s high, low and close to calculate the next session’s classic pivots. PP = (H + L + C) / 3; R1 = 2PP − L; S1 = 2PP − H.
AFFLATUS
TRADING ARENA
Full S&P 500 TA dashboard · Autopilot vs SPY/SMH
TRAXUS//CVKM · simulation deck
Observe markets. Review simulations.
Three simulated LLM ledgers trade the S&P 500 under fixed risk rules and are scored against SPY/SMH. Review the latest published snapshot, or search any ticker for daily technicals. Historical simulation only — not investment advice.
Mode, exchange session and data freshness are independent. This page sends no orders and uses no real money.
- Feed
- UNKNOWN · quote snapshots; entitlement and delay unverified
- Execution
- NONE · intraday Paper unavailable
- Research clock
- Daily bars: completed sessions. Ledger: published record timestamps (UTC).
—
NEWS · —
QUOTA KEY SET
US MARKET OPENS IN
--:--:--
NYSE core equities · holidays and early closes apply
—
Published watchlist
Symbols come from the published quote allowlist. Select one to inspect available daily data, independently of archived proposals below.
Archived proposals and reasons
Inspect source snapshot (JSON)Select a watchlist symbol or search to inspect daily data.
Ticker calculations require JavaScript and available daily history. Read both reference methods below, or inspect the published source data.
POST-MARKET · review the last session
Calculate reference pivots from the preceding session, then compare them with the latest completed session’s range and close to review tested levels. Missing daily data cannot establish a test or breakout.
Decision & execution timeline
Published fills and risk rejections, newest first. These records are not an intraday replay engine. Missing decision inputs remain unavailable.
Published Market Signal
…
Q-FOUNDRY // INDEPENDENT RESEARCH ENGINE
A risk engine, not another stock picker.
QF-01 turns the current AI-factory research universe into a reproducible walk-forward experiment: completed daily bars enter a time-bounded data context, four cross-sectional factors become an explainable score, and hard portfolio constraints decide what survives.
QF-01
manifest · —
source commit · —
01DATA CONTEXTcompleted sessions
02FACTOR LENSrobust cross-section
03RISK REGIMEtrend × volatility
04CONSTRAINT GATEname × sector caps
05WALK-FORWARDcosted out-of-sample
RESEARCH UNIVERSE · NO PRINCIPAL / NO ORDERSSPY // BENCHMARK
CONTEXTFACTORSPORTFOLIORISK
Compile when you want the browser to request the latest completed daily histories. Nothing here is annualized from a single trade, no future bar enters a signal, and every rebalance pays the configured cost.
REGIME—
GROSS—
β / SPY—
WINDOW—
QF-01costed walk-forward
QF-01 SPY
CONSTRAINED PORTFOLIOweight is normalized exposure, never account value
| ASSET | SCORE | MOM | VOL | β | WEIGHT |
|---|
OPEN FACTOR DIAGNOSTICS
| ASSET | MOM | TREND | RESIL. | LOW VOL | SCORE |
|---|
LOCAL ITERATION TAPEdevice-only · newest first
- —
WHY THIS IS BASED ON GS-QUANT
The implementation follows the repository's architectural seams, not its proprietary services: an explicit time context, a portfolio as a first-class object, factor and specific-risk diagnostics, then an optimizer gate with position, sector and turnover constraints. Arena replaces Marquee-bound calls with public completed-session data and a transparent local engine.
FeidaWang/gs-quant ↗
DataContext ↗
Portfolio ↗
FactorRiskModel ↗
OptimizerConstraints ↗
GenericEngine ↗
Independent educational research, not affiliated with or endorsed by Goldman Sachs. Backtests are hypothetical, sensitive to assumptions and transaction costs, and are not investment advice.