Probabilistic Principal Component Analysis


Principal component analysis (PCA) is a ubiquitous technique for data analysis and processing, but one which is not based upon a probability model. In this paper we demonstrate how the principal axes of a set of observed data vectors may be determined through maximum-likelihood estimation of parameters in a latent variable model closely related to factor analysis. We consider the properties of the associated likelihood function, giving an EM algorithm for estimating the principal subspace iteratively, and discuss, with illustrative examples, the advantages conveyed by this probabilistic approach to PCA.

Extracted Key Phrases

3 Figures and Tables

Showing 1-10 of 13 references

Multivariate Analysis Part 2: Classification, Covariance Structures and Repeated Measurements

  • W J Krzanowski, F H C Marriott
  • 1994
1 Excerpt

Statistical Factor Analysis and Related Methods

  • A Basilevsky
  • 1994
2 Excerpts

The Statistical Analysis of Finite Mixture Distributions

  • D M Titterington, A F M Smith, U E Makov
  • 1985
1 Excerpt

EM algorithms for ML factor analysis

  • D B Rubin, D T Thayer
  • 1982
1 Excerpt

Asymptotic theory for principal component analysis

  • T W Anderson
  • 1963
1 Excerpt

Statistical inference in factor analysis

  • T W Anderson, H Rubin
  • 1956
2 Excerpts

A modified method of estimation in factor analysis and some large sample results

  • D N Lawley
  • 1953
2 Excerpts
Showing 1-10 of 1,037 extracted citations
Citations per Year

1,999 Citations

Semantic Scholar estimates that this publication has received between 1,784 and 2,240 citations based on the available data.

See our FAQ for additional information.