Statistical Power
Also called: power of a test, underpowered study
Statistical power is the probability of detecting an effect that genuinely exists — one minus the Type II error rate. Underpowered studies are the quiet epidemic of empirical research: with too small a sample, real effects go undetected, and the few results that clear significance are exaggerated by construction. In finance this is why decades of data are needed to distinguish a genuinely skilled manager from a lucky one.
Formula
Power = 1 − P(Type II error)
Want more than a definition? Learn it in Stat Dojo →