- execution/trader: a held US stock with no print for 5 sessions that has left
the index is sold at its last print (execution.SETTLE_AFTER_SESSIONS), in both
the batch replay and the live session ledger. Acquired names used to stay at
their last mark forever: lowrisk_top20 ended 2017-12..2026-10 with 11 of 26
dead positions, and CAGR rises 1.7-5.5pp once they settle. ENGINE_VERSION
cash-ledger-pit-v3; the API cache key follows.
- universe_history: current members whose pre-2011 addition the Wikipedia change
log missed now start at the list page's "Date added" (80 of 94 current members
added 2001-2010 were open-start, e.g. ISRG, MA, BKNG). Membership from 2016 on
is unchanged; the research panel cache now keys on the membership intervals.
- research/lab/registered: pure-ETF strategies replay on the ETF panel (they ran
as all-cash books or crashed); prepare_strategy refuses an ETF strategy that
lacks its ETF universe.
- research/lab/factors: trend_hzz uses a month's regression betas only from the
session its forward returns end (short months read 1-2 sessions ahead).
- tests: every registered strategy (US and CN) is truncation-invariant on the
production signal path; batch and live ledgers settle a delisting identically.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Share point-in-time signals and a cash-funded order ledger across backtests and trading. Correct eligibility, missing-price fills, quality factors, regime inputs, warmup windows, and cash-injection accounting. Add reproducible registered-strategy comparisons and paired forward candidates with a 20% resize band.
Validation: 125 Python tests, 23 web tests, TypeScript/Vite build, 87 strategy causality checks, and real-data target/order/equity parity for 10 strategies.
Strategies
- BarStrategy protocol (close/open/high/low/volume bars; row d uses bars
through d-1) and RobustMomentumStrategy: overnight_recovery_top10 (US),
lowrisk_momentum_top15 and lowbeta_sharpe_mom_top20 (CN), top-N with
rank hysteresis, 10-day rebalance, weights drift between rebalances.
- trader: strategy_fields/strategy_bars/bar_session_weights; one signal
path (_signals_for_sessions) replaces generate_target_weights.
Live fixes
- CSI 300 constituents come from China Securities Index's file; the
Wikipedia table was 122 names off the real index.
- S&P 500 change log moved to "Historical components of the S&P 500".
- data_manager: tickers missing from a partial Yahoo batch are retried
alone before being marked unavailable (CEG, GEV, SNDK ... were dropped).
update() takes fields=[...] and stores high/low/volume panels.
- Whole-lot orders go to the nearest lot within cash (CN books held ~62%
cash); held names are resized only beyond RESIZE_BAND (10%), so books no
longer pay a ticket fee to trade daily drift.
Web backtests
- backtest_runner.py replaces the duplicated bridge/precompute code: PIT
membership in both markets (data/csi300_history.json), stock universe
limited to index members (ETFs and benchmarks in the shared panel were
pickable), and the book kept like the live trader (simulate_live_book:
signal timing, drift, order floor, resize band, fees on orders placed).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
- Monthly/yearly returns chain from the previous period's close (the first
trading day of every period was dropped; up to ~9pp/yr error).
- With cash injections, metrics/drawdowns/periodic returns are time-weighted
on the no-injection curve; capitalFlows reports invested/profit separately.
- Benchmarks cover exactly the strategies' dates.
- CAGR uses calendar time, so gaps in live records do not inflate it.
- Backtests skip the per-trade fee for rebalances below the live minimum
order and keep a ruined portfolio at zero.
- live: exact strategy match, ineligible strategies hidden, currency,
startDate, same-window benchmark, stale-price fallback, equity anchored at
the initial capital.
- PIT membership falls back to the cached history when the Wikipedia change
log cannot be parsed (the table is no longer on the page).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The existing framework fetches today's S&P 500 constituents from Wikipedia
and applies that list to the entire 10-year price history — classic
survivorship bias. Stocks that went bankrupt or were removed for poor
performance are absent, while today's winners (which may have been minor
names 10 years ago) are implicitly selected. This materially inflates
reported strategy returns.
New pipeline:
- universe_history.py reconstructs per-ticker membership intervals by
walking Wikipedia's "Selected changes" table backward from today.
- research/fetch_historical.py downloads prices for all 848 tickers
that were ever members (Yahoo returns ~675 of them; ~170 fully
delisted names are unavailable — remaining partial bias).
- research/pit_backtest.py masks prices to NaN outside membership
windows so strategies naturally cannot select non-members.
- research/strategies_plus.py adds RecoveryMomentumPlus (generalized
Recovery+Momentum with configurable weighting / blend / regime hook)
and an EnsembleStrategy.
- research/optimize.py runs five experiments: bias drift, hyperparameter
sweep (2016-2022 train / 2023-2026 test), SPY MA regime filter,
weighting schemes, and an uncorrelated-config ensemble.
Headline finding: the biased backtest reports 40.9% CAGR for
recovery_mom_top10 over 2016-2026; the point-in-time version reports
22.4% (vs 14.0% SPY buy-and-hold). True edge is ~8pp CAGR, not ~27pp.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
