Credible Interval

A credible interval is the Bayesian counterpart to a confidence interval: a range that, according to your posterior, contains the true value with a stated probability. A 95% credible interval lets you honestly say “given my data and prior, there’s a 95% probability the value lies in here.”

This is exactly the intuitive statement a confidence interval is so often misread to mean — and the difference is real, not pedantic. The credible interval’s direct probability comes at the cost of depending on the prior; the confidence interval avoids a prior but only makes a long-run guarantee about the procedure, not about your one interval.