Flat prior
Topic of Conjugate analysis for normal models, part of Bayesian statistics
Flat prior
| Type | Topic |
| Taught in | Conjugate analysis for normal models |
| Part | Bayesian statistics |
| Questions using it | 3 |
A prior density that is constant in the parameter. On an unbounded parameter it is improper. The posterior is then proportional to the likelihood, so its mode is the maximum likelihood estimate; a flat prior is flat only on the scale chosen, not on transformations of it.
In Conjugate analysis for normal models
A proper approximation to the flat prior
Used in
- Taught in Conjugate analysis for normal models
- Also used in Bayesian foundations · The Beta–Binomial model · Categorical data and the Dirichlet distribution · Bayesian hypotheses and model comparison · Hierarchical Bayes
- Questions that use it (3) — A geometric observation and a uniform prior (answer) · Q11 (answer) · Q58
- All 19 questions on Conjugate analysis for normal models
See also
Likelihood function · Maximum likelihood · A proper approximation to the flat prior · Maximum a posteriori estimate · Normal–Normal model · Normal–Normal conjugacy · Precision weighting · Inverse-Gamma distribution