form4 dollar weighted cluster
What it checks
Insider buy clusters are most predictive when the dollar amount is large relative to the firm's own history, not relative to an absolute threshold. Rank each company's 30-day rolling insider dollar-buy against its 1-year history and trade the top-decile events.
Mechanism
Existing #31 insider_form4 and form4_cluster_anomaly combine count + dollar thresholds but neither uses the per-ticker percentile rank of rolling-30d $-buy as the primary signal. Filters out richly-followed names where small-dollar transactions are routine, and elevates small/mid-cap conviction signals that would otherwise miss the $1M floor.
Signal rule
30-day rolling $-buy in top decile (>=P90) or top vigintile (>=P95) of own 252d history + 60d uptrend fires LONG 60/120/180d.
Data dependencies
daily_pricesAdjusted-close OHLCV for every US-listed ticker; primary price feed.
sec_insider_tradesForm-4 insider transactions with role, size, and trade direction.
Expected edge
- Paper alpha
- +4-6% over 6mo (vs count-only cluster)
- Paper window
- T+1 to T+180d
Cohen-Malloy-Pomorski 2012 (extension); +4-6% over 6mo on $-weighted top-decile vs equal-weighted cluster.
Related families
insider form4Insider & FlowLakonishok-Lee 2001, Cohen-Malloy-Pomorski 2012: corroborated insider purchases (Form 4 transaction_code='P') from multiple insiders within a short window are the single most-predictive insider signal โ ~6%/year alpha over 6-12 months. Single-insider buys are noisy; cluster buys (3+ distinct insiders, or a single buy > $1M from CEO/CFO) are not. Long for 60 days on cluster buys; long for 30 days on a single $1M+ transaction. Sales are deliberately ignored โ they're confounded by liquidity, diversification, and options expiry.
form4 cluster anomalyInsider & FlowAnomaly-tail Form 4 insider PURCHASE clusters (>=4 distinct insiders within 30 days, or >=3 insiders with aggregate dollar value >= $2M) strip routine-liquidity noise that dominates single-insider buys. Long the issuer 60-180d post-cluster.
cluster buy post drawdownInsider & FlowCohen-Malloy-Pomorski 2012 document that the strongest insider-purchase drift comes from non-routine clusters, and the subset where insiders are buying AFTER a price drawdown is the highest-information leg. A routine insider buys in all regimes; a conviction insider buys when prices are down. Drawdown-conditioned clusters strip out the routine noise that dominates single-insider filings.
Explore form4 dollar weighted cluster on alphactor.ai
See which tickers this family is currently firing on, with live signals and rankings.