Bayesian Inference
Also called: Bayesian statistics, prior and posterior, Bayesian updating
Bayesian inference treats unknown parameters as having probability distributions, starting from a prior and updating to a posterior as data arrives. It produces statements people actually want — 'a 90% chance the parameter lies here' — which frequentist confidence intervals technically do not. The trade-offs are choosing a defensible prior and the computational cost, which is why the approach only became practical with modern sampling methods.
Want more than a definition? Learn it in Stat Dojo →