Probability & Statistics Codexery

Posterior probability

Last updated

Bayes' rule is used to combine a prior probability with information from the likelihood, producing a conditional probability known as the posterior probability.[1] Because it depends on randomly observed data, the posterior probability qualifies as a random variable.[2]

Quick Facts

Example result
In a school with 60% boys and 40% girls, probability a trouser-wearing student is a girl is 25%

Facts from the source article.

Key facts and history

Bayes' rule provides a way to compute conditional probabilities, such as the probability of having a disease given a positive test result.[3] In medical testing, the false positive rate is the probability of a positive result conditional on the person having no disease.[3] A prior and likelihood are conjugate when the resulting posterior distribution is the same type of distribution as the prior.[4] For binomial data, using a beta prior yields a beta posterior, and the posterior parameters are Beta(α+x, β+n-x).[4]

The order or frequency of adding data batches does not affect the final posterior.[4]

What the research sources say

Clyde explains that Bayesian statistics mostly involves conditional probability, which can be calculated using Bayes' rule, and is widely used in medical testing.[3] Etz describes likelihoods as the bridge that gets us from prior to posterior in Bayesian inference.[4] Etz illustrates how different priors can converge into highly similar posteriors using data from an NCAA 3-point shooting contest.[4] When a prior has only 1 or 2 data points worth of information, it has essentially no impact on the posterior shape.[4]

The breadth of posterior curves can be thought to represent the uncertainty in estimates.[4] Clyde notes that a Bayesian credible interval is a range for which the Bayesian thinks the probability of including the true value is, say, 0.95.[3] Clyde provides an example where a posterior distribution yields a 95% credible interval of 60% to 64% for a population proportion.[3]

References

  1. Wikipedia, "Posterior probability" (CC BY-SA 4.0), which cites: Lambert, "A Student's Guide to Bayesian Statistics", Sage, 2018, p. 121–140; Grossman, "Inferences from observations to simple statistical hypotheses", University of Sydney, 2005; Etz, "Understanding Bayes: Updating priors via the likelihood", The Etz-Files, 2015-07-25 (we also read that source) en.wikipedia.org
  2. Wikipedia, "Posterior probability" (CC BY-SA 4.0), which cites: Clyde, "Chapter 1 The Basics of Bayesian Statistics An Introduction to Bayesian Thinking" (we also read that source) en.wikipedia.org
  3. Clyde, "Chapter 1 The Basics of Bayesian Statistics An Introduction to Bayesian Thinking". facts only, no text reused. statswithr.github.io
  4. Etz, "Understanding Bayes: Updating priors via the likelihood", The Etz-Files, 2015-07-25. facts only, no text reused. alexanderetz.com

Each numbered sentence above was checked against the source it cites before publication. Sentences citing Wikipedia adapt text from the English Wikipedia article Posterior probability (contributors), used under CC BY-SA 4.0; this entry is available under the same licence. Other sources were used for facts only; no text or images are copied from them.

More in Probability & Statistics

Sources

Compiled from Wikipedia and the sources listed below. Text from Wikipedia is available under CC BY-SA 4.0; this entry is adapted from it.

Spotted an error? Know more?

Reader corrections go straight into our review queue. Suggest an edit · How this site is sourced

Comments

Loading…
Open in the interactive codex →