Skip to content

Extracted algorithm research record

OnlineLogReg

Online logistic regression with SGD for adaptive trade signal filtering.

Record 497 · Catalog rank 967 · Statistical, Quantitative & Machine Learning · Subtype I1 · Snapshot 2026-10-31 15:46 UTC

Source & extraction context

Open original source ↗

Evidence label: backtest_shown (source claim, not independent verification). Recorded completeness: 50.0/100. Neither is a performance result.

Names identify source records. Inclusion is not an endorsement or a claim of authorship, ownership or reuse permission. Check the source license before reusing any implementation.

Extraction warnings

Opacity warnings

Recorded market & execution profile

Recorded markets
any
Recorded timeframes
M15
Recorded sessions
any
Recorded regime
any

Recorded indicators & parameters

Extracted indicator stack
IndicatorParametersApplied to
EMA{"period": 12, "period2": 26}close
RSI{"period": 14}close
ATR{"period": 14}close
MACD{"fast_period": 12, "signal_period": 9, "slow_period": 26}close
SMA{"period": 20}volume

Extracted strategy rules

Recorded long-entry rules

Recorded short-entry rules

Recorded exit rules

Recorded filters

Stop loss
ATR multiple, 1.5 by default
Take profit
ATR multiple, 2.5 by default
Position sizing
Fixed-fractional off the ATR-based stop distance, rounded to the broker's lot step and clamped to min/max volume

Questions about this record

What entry rules are recorded for OnlineLogReg?

Long: Predict(feat) < InpProbThreshold Short: ... These are unverified extracted rules.

Which markets and timeframes are recorded?

Markets: any. Timeframes: M15. These are source-recorded labels, not verified execution behavior.