Prior and Posterior Interview Questions
Prior and posterior interview prep for Bayesian updates, evidence, likelihoods, count tables, and base-rate mistakes.
Candidates practicing Bayesian probability and signal-update prompts.
Prior comes before evidence
A prior is the probability you assign before seeing the new signal or evidence. It is not optional in Bayes-style questions.
Posterior comes after evidence
A posterior is the updated probability after accounting for the evidence. It should reflect both the prior and how likely the evidence is under each state.
Concrete example
If a rare state has prior probability 1 percent, even a fairly accurate signal can produce many false positives. The posterior must compare true-signal and false-signal counts.
Likelihood role
The likelihood measures how compatible the evidence is with a state. It updates the prior but does not replace it.
Use count tables
A count table turns priors and likelihoods into concrete groups. This often prevents base-rate mistakes.
Common mistakes
Candidates often quote the signal accuracy as the posterior. The posterior is the probability of the state after the signal, not the signal accuracy.
Practice the pattern
Use the LeetQuidity curriculum and calibration to turn this topic into a focused practice plan.