AI · SG REITs — Trades
Every virtual fill this book has made, with the model's reasoning — the fused signal, valuation and news case — behind each. Prices are model fills; quantities are never shown.
Fill history
1 recorded fill.
| Date | Ticker | Name | Side | Fill Px | Signal | Conv | Valuation | Why |
|---|---|---|---|---|---|---|---|---|
| 2026-08-03 | N2IU | Mapletree PanAsia Com Tr | BUY | 1.33 | PullbackBuy_VWAP | 86 | GROWTH PRICED IN | rationaleMapletree Pan Asia Commercial Trust is the first Singapore REIT in this book's history to satisfy every one of our entry rules at once, and we are buying it the patient way. The trust has now signalled a buy twice in a row on the same pattern we trust most - a pullback to its volume-weighted average price - strengthening from a moderate reading to the strongest in weeks, with price above every major moving average and the trend firmly established. Crucially, we are not chasing it: today's close ran 2.2% ahead of the price we are willing to pay, so we are leaving a resting order below the market that fills only if the shares come back to us. That structure is deliberate, because the US Federal Reserve decides on rates tonight and the trust reports quarterly numbers tomorrow. If the market rallies away from us we simply do not get filled and reassess; if it dips, we buy at a better price with a stop just 2.3% lower. The valuation screen still marks this trust down, as it does every distribution-paying REIT, but the signal now clears our bar even after that penalty is applied in full - the first time that has happened. OPERATIONAL: SGR-3 satisfied (2nd consecutive same-family PullbackBuy_VWAP print, both >= 65). SGR-2 clears (T12 1.39 >= close 1.36, +2.2%). L1-favoured family. No-chase: Entry_Distance_ATR -1.14, buy-limit only. Floor cleared on the FULLY PENALISED arithmetic (96 - 10 + 0 = 86); Quality_Pass=False hard gate OVERRIDDEN under SGR-1 (REIT net-income-EPV/F-Score screen known-miscalibrated; second time this override has enabled an acquire, after C38U 07-23). Size 0.5x on STAGING grounds (FOMC 07-30, MPACT 1Q 07-30, PMI 08-03, sector 1H 08-05), not on the valuation modulator, which SGR-1 nulls. Sizing: 96.51% x Sown seed (ADR-0014) x 20% buy_pct = 386,040; max_position_pct 60% no bind; x0.5 = 193,020; /1.33 = 145,127 -> lot 100 -> = S = 9.65% of book. Risk if stopped: S = 0.22% of book. L8: executes at the 07-30 open as a resting GTC limit; reprice/re-validate if the open gaps >1.5%; expires_on 2026-08-03 (3 sessions). ANNOTATIONS: 12wk 1.39 exceeds DCF anchor 1.1313 (GROWTH_PRICED_IN) - projection assumes growth beyond priced-in; and 12wk 1.39 < 4wk 1.42 is internally incoherent (see systemic finding 4) - treat 1.42 as the trade objective, 1.39 as the softer figure. f=0.30 on the 1wk per L5 (calib_1w null, 1 graded entry). Withdraw if conviction < 65 at any subsequent EOD. · news decided @ 1.33 · stop 1.30 · t1w 1.36 · t4w 1.42 · t12w 1.39 |
Learnings in force
What this book's own record has taught it.
Last reviewed 2026-09-13
Favor PullbackBuy_VWAP entries over breakout/FailedBreakdown chases.
`PullbackBuy_VWAP|buy` is the most reliable buy cell (US hit ~0.86); `FailedBreakdown_Up_20D|buy` chases are weak (~0.25, stop-first ~60%) and `BreakoutUp_52W` 4wk targets overshoot +55-75% and mean-revert. On ties prefer the PullbackBuy; discount breakout/FailedBreakdown chases.
Trim on the stop-break + negative-12wk combo; don't wait for the SELL-REDUCE label.
A decisive close below the recorded stop AND a negative 12wk is itself a signal-based exit (CLAUDE.md §9), even while the engine still prints NO TRADE.
Execute the de-risk promptly.
A trim recommended but not executed keeps bleeding; act when a held name closes below stop with a negative 12wk, don't re-recommend the same trim for days.
The settled close is authoritative; don't act on weak opens.
Intraday/BOD prints whipsaw. Flag intraday reads provisional; reserve firm acquire/dispose for the settled-close EOD.
Distrust the model 1-week estimate when `calib_1w` is degenerate.
If the scorecard `calib_1w_fraction` is <=0 or wildly unstable, fall back to f=0.30 for the 1wk interpolation and flag it.
Don't chase post-earnings gaps.
A blowout (e.g. MU +14% AH) often round-trips within days; wait for the pullback to the model entry.
(ARCHIVED for sgreit — ADR-0014 standalone) Same-pool/cross-book shares execute ONCE.
This lesson applied when sgreit mirrored `sg`. Post-2026-07-06 sgreit runs on its own S seed cash with no shared shares, so there is nothing to reconcile here; this book's own cash independently. Retained only as the generic rule for any book that still genuinely shares a pool with `sg`.
EOD recommendations execute at the NEXT market open, not the decision close.
Persist actions as `execution_status: pending_next_open` with an `execute_on` date; action prices are close-refs — reprice/re-validate if the open gaps >~1.5%; the 1wk/4wk/12wk horizons and the scorecard grade from the execution date/price; live holdings/monitor stay unchanged until the JSON is updated post-fill; the next BOD reconciles pending proposals against the actual open.