Using Model Selection Criteria to Choose the Number of Principal Components
Stanley L. Sclove
Abstract
Abstract The use of information criteria, especially AIC (Akaike’s information criterion) and BIC (Bayesian information criterion), for choosing an adequate number of principal components is illustrated.
Topics & Concepts
Akaike information criterionBayesian information criterionMathematicsInformation CriteriaStatisticsSelection (genetic algorithm)Deviance information criterionPrincipal component analysisModel selectionPrincipal (computer security)Bayesian probabilityBayesian inferenceComputer scienceArtificial intelligenceOperating systemStatistical Methods and InferenceAdvanced Statistical Methods and ModelsBayesian Methods and Mixture Models