Gradient Boosting for Insider Purchase Signals in Microcap Stocks

Summary

The paper tests whether U.S. microcap companies’ open-market insider purchases, reported through regulatory disclosures, can help predict subsequent abnormal returns. It analyzes 17,237 purchases involving 1,343 issuers from 2018 to 2024, with market capitalizations from $30 million to $500 million. A gradient boosting model uses insider characteristics, prior transaction behavior, and market conditions available at disclosure. On held-out 2024 data, the classifier records an AUC of 0.70; at a selected threshold, precision is 0.38 and recall is 0.69.

The strongest reported predictor is how far a stock sits below its 52-week high. Purchases disclosed after gains exceeding 10% have the largest mean cumulative abnormal return and outperformance rate in the analysis, suggesting a trend-confirmation pattern rather than a simple reversal effect. The authors report that the results persist after winsorization and across subsamples. The findings are predictive associations, not proof that insider buying causes returns; they are specific to the studied microcap range and period, and illiquidity may affect both signal behavior and practical execution.

Key ideas

• The study evaluates insider purchase disclosures as predictors of abnormal returns in U.S. microcap equities.
• A gradient boosting classifier combines insider history and characteristics with market conditions at disclosure.
• Distance from the 52-week high is the model’s most influential reported feature.
• Purchases disclosed after substantial price gains show stronger observed performance than the study’s reversal intuition would suggest.
• The reported patterns survive winsorization and appear across subsamples, though they do not establish causation or guarantee tradable returns.

Tags

# Insider Purchase Signals in Microcap Equities: Gradient Boosting Detection of Abnormal Returns


# Insider Purchase Signals in Microcap Equities: Gradient Boosting Detection of Abnormal Returns









This paper examines whether SEC Form 4 insider purchase filings predict abnormal returns in U.S. microcap stocks. The analysis covers 17,237 open-market purchases across 1,343 issuers from 2018 through 2024, restricted to market capitalizations between \$30M and \$500M. A gradient boosting classifier trained on insider identity, transaction history, and market conditions at disclosure achieves AUC of 0.70 on out-of-sample 2024 data. At an optimized threshold of 0.20, precision is 0.38 and recall is 0.69. The distance from the 52-week high dominates feature importance, accounting for 36% of predictive signal. A momentum pattern emerges in the data: transactions disclosed after price appreciation exceeding 10% yield the highest mean cumulative abnormal return (6.3%) and the highest probability of outperformance (36.7%). This contrasts with the simple mean-reversion intuition often applied to post-run-up entries. The result is robust to winsorization and holds across subsamples. These patterns are consistent with slower information incorporation in illiquid markets, where trend confirmation may filter for higher-conviction insider signals.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.