Loading gamma flip levels…
Back to GlossaryChartGEX Methodology

Bonferroni Correction

A statistical correction that raises a detection threshold proportionally to the number of independent hypotheses being tested. ChartGEX applies a correlation-adjusted Bonferroni correction to dealer-structure detection across strike and expiration scans.

Also known as:multiple comparisons correctionfamily-wise error rate

The Bonferroni correction adjusts statistical significance thresholds when multiple hypotheses are tested simultaneously. If you scan 100 independent strikes for an anomaly with a 5% individual significance threshold, you would expect 5 false positives by chance alone. The Bonferroni correction raises the threshold proportionally to keep the family-wise error rate at 5% across the entire scan.

ChartGEX's correlation-aware variant

Adjacent strikes in SPY options have correlation coefficients above 0.9 — they are not independent hypotheses. A naive Bonferroni correction over hundreds of strikes would be far too conservative, suppressing real signals. ChartGEX applies a 6× correlation factor for SPY: instead of dividing by the full strike count, it divides by the effective number of independent tests after accounting for cross-strike correlation. The result keeps the false-positive rate disciplined without burying real dealer-structure events under a wall of multiple-comparisons penalty.

This is one of several design choices that reflect ChartGEX's commitment to methodology transparency: signals are not just labeled, they are quantitatively defended against statistical noise.