What is a Credible Interval?
You'll learn: What a credible interval is and how it differs from traditional confidence intervals through a Bayesian lens.
The Core Idea
A credible interval is a Bayesian probability statement that tells you: "Given the data I've observed, there's a certain probability (like 95%) that the true parameter value lies within this range."
Unlike frequentist confidence intervals (which make statements about hypothetical repeated experiments), credible intervals make direct probability statements about where we believe the parameter actually is, right now, based on the data we have.
A Direct Probability Statement
After computing your posterior distribution using Bayes' theorem, a 95% credible interval contains 95% of the posterior probability mass. You can legitimately say: "There's a 95% probability the parameter falls in this interval, given the observed data."
This is the natural interpretation most people want from an interval estimate—and Bayesian analysis delivers it directly.
How It Works
- You start with a prior distribution (your initial beliefs)
- You collect data and compute the likelihood
- You calculate the posterior distribution
- You find the interval that captures your desired percentage (e.g., 95%) of the posterior probability
Example Interpretation
If your 95% credible interval for a conversion rate is [0.12, 0.18], you can say: "Given our data, there's a 95% probability the true conversion rate is between 12% and 18%."
This contrasts with frequentist language, which cannot make probability statements about the parameter itself.
Key Takeaway: A credible interval is a Bayesian probability statement about where a parameter likely lies given observed data—it directly quantifies uncertainty about the parameter's value using the posterior distribution.