Abstract | ||
---|---|---|
ABSTRACT Recently, the technique of principal component analysis (PCA) has been expressed as the maximum,likelihood solu- tion for a generative latent variable model. A central issue in PCA is choosing the number,of principal components,to re- tain. This can be considered as a problem of model selection. In this paper, the probabilistic reformulation of PCA is used as a basis for a Bayasian approach of PCA to derive a model selection criterion for determining the true dimensionality of data. The proposed criterion is similar to the Bayesian Infor- mation Criterion, BIC, with a particular goodness of fit term and it is consistent. A simulation example,that illustrates its performance,for the determination of the number,of principal components,to be retained is presented. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1016/j.sigpro.2006.09.001 | Signal Processing |
Keywords | Field | DocType |
bayes methods,principal component analysis,bic,bayesian estimation,bayesian information criterion,pca,generative latent variable model,maximum likelihood solution,probabilistic reformulation,model selection,mathematical models,goodness of fit,maximum likelihood estimation,principal component,estimation,information theory,latent variable model,maximum likelihood | Sparse PCA,Bayesian information criterion,Pattern recognition,Latent variable model,Bayes factor,Model selection,Bayesian network,Artificial intelligence,Goodness of fit,Principal component analysis,Mathematics | Journal |
Volume | Issue | ISSN |
87 | 3 | Signal Processing |
Citations | PageRank | References |
6 | 0.68 | 1 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Abd-Krim Seghouane | 1 | 193 | 24.99 |
Andrzej Cichocki | 2 | 5228 | 508.42 |