Missing Value Estimation Using Mixture of PCAs

Shigeyuki Oba,Ken-Ichi Matsubara,Ichiro Takemasa,Masa-Aki Sato,Shin Ishii,Morito Monden

doi:10.1007/3-540-46084-5_80

Missing Value Estimation Using Mixture of PCAs

Shigeyuki Oba, Ken-Ichi Matsubara + Show 4 more

https://doi.org/10.1007/3-540-46084-5_80

Copy DOI

Publication Date: Jan 1, 2002

Citations: 7

Affiliation: Nara Institute of Science and Technology, Osaka University, Research Organization of Information and Systems, Japan Science and Technology Agency

#Variational Bayes Method #Mixture Of Principal Component Analyzers + Show 8 more

Abstract
Full-Text PDF
Similar Papers

Abstract

We apply mixture of principal component analyzers (MPCA) to missing value estimation problems. A variational Bayes (VB) method for MPCA with missing values is developed. The missing values are regarded as hidden variables and their estimation is done simultaneously with the parameter estimation. It is found that VB method is better than maximum likelihood method by using artificial data. We also applied our method to DNA microarray data and the performance outweighed the conventional k-nearest neighbor method.

Full Text