Bayesian information criterion
Definition taught in Model selection for stochastic models, part of Statistical learning
Bayesian information criterion
| Type | Definition |
| Taught in | Model selection for stochastic models |
| Part | Statistical learning |
| Questions using it | 5 |
Multiply the approximate MAP rule by -2, which reverses the direction of the optimisation. The same rule is
k^*\approx\arg\min_k \mathrm{BIC}_k, \qquad \mathrm{BIC}_k=-2\ell(\widehat\Psi_k;X_{\mathrm{all}},M_k)+d_k\log n.
Used in
- Taught in Model selection for stochastic models, where the derivation and examples are.
- Also used in Statistical classifiers and discriminant analysis · Logistic regression · High-dimensional analysis · Graphical models · Forecasting
- Questions that use it (5) — The within-cluster sum cannot choose K (answer) · BIC and AIC by hand · A tempting wrong turn (answer) · Choosing model complexity · Mixture selection and uncertainty
- All 7 questions on Model selection for stochastic models
See also
Marginal likelihood and Bayes factor · Bayes factors and posterior odds · Akaike information criterion · Extended information criterion · Maximum a posteriori estimate · Laplace approximation · Dropping terms that do not scale with n · Reading the BIC