🏠
Guest Not signed in

Law of total probability: average over the hidden case

Split the world into mutually-exclusive cases, compute the conditional in each, weight by the case probability, sum.

Method · Law Total Probability
Intro

When a probability depends on an upstream choice you can’t see, condition on that choice. The law of total probability says: if $B_1, \ldots, B_n$ partition the sample space, then $P(A) = \sum_i P(A \mid B_i)\, P(B_i)$ — the marginal is a weighted average of the conditionals, with the priors as weights. It’s the denominator of Bayes’ rule and the workhorse for any “case splits” problem.

βœ“ Intro Β· expand
Independent · Legal