Quantitative Research

Test 1 — Is it just market-crash reversion?

Insider buying clusters on certain days — the same days markets are panicking (COVID March 2020, the 2008-09 crisis, SVB 2023). If the "signal" were just broad market reversion after a crash, restricting to calm days should weaken it.

The trade journal is barely profitable at a token $200 stake per trade. Before concluding the effect is too small to matter, three ways it could be a statistical illusion were tested directly against the raw data — a crash-clustering artifact, inflated significance from correlated same-day events, and pure bid-ask bounce in illiquid names. All three were built to make the finding disappear. None of them did.

The opposite happens. The t-statistic is stronger on ordinary days than on high-clustering days — a crash-reversion artifact would show the reverse.