Rejected thesis
Insider cluster buying
Thesis (UNTESTED, blocked on data): When 3+ insiders independently buy within a 30-day window totalling >= $500k, the stock outperforms over the next 30-90 days (Cohen / Malloy / Pomorski 2012).
Code: research/insider_cluster_validate.py
Report: research/insider_cluster_report.json (n=0)
Why it failed: The insider_transaction table in db/finance.db
has only 44 purchase rows across 19 symbols. The Form-4 ingestion
(scripts/build_insider.py) hasn't been run across the universe.
Validator is ready to run once the backfill is done.
Verdict: Re-test after running python -m scripts.build_insider. The
thesis has strong academic priors and the validator is wired up; this is
a "blocked on operational work" rather than a real failure.
| Sample | 0 |
|---|---|
| Validator | research/insider_cluster_validate.py |
| Verdict | Rejected |
Every result here is reproducible from the script named above. Reports are in the repository.
- Volatility squeeze
- Pre-FOMC drift
- Uptrend pullback
- Momentum continuation
- Industry-relative momentum (at 4-week lookback)
- Earnings-reaction reversal (anti-PEAD)
- Beat-persistence pre-earnings drift
- Sector ETF mean-reversion
- 52-week-high breakout
- Volume accumulation (above-200DMA +...
- Cross-sector momentum
- Analyst-upgrade cluster