Dimensionality reduction for binary data through the projection of natural parameters

DOI10.1016/J.JMVA.2020.104668zbMATH Open1450.62069arXiv1510.06112OpenAlexW2164214353MaRDI QIDQ2201560FDOQ2201560

Yoonkyung Lee, Andrew J. Landgraf

Publication date: 29 September 2020

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Abstract: Principal component analysis (PCA) for binary data, known as logistic PCA, has become a popular alternative to dimensionality reduction of binary data. It is motivated as an extension of ordinary PCA by means of a matrix factorization, akin to the singular value decomposition, that maximizes the Bernoulli log-likelihood. We propose a new formulation of logistic PCA which extends Pearson's formulation of a low dimensional data representation with minimum error to binary data. Our formulation does not require a matrix factorization, as previous methods do, but instead looks for projections of the natural parameters from the saturated model. Due to this difference, the number of parameters does not grow with the number of observations and the principal component scores on new data can be computed with simple matrix multiplication. We derive explicit solutions for data matrices of special structure and provide computationally efficient algorithms for solving for the principal component loadings. Through simulation experiments and an analysis of medical diagnoses data, we compare our formulation of logistic PCA to the previous formulation as well as ordinary PCA to demonstrate its benefits.

Full work available at URL: https://arxiv.org/abs/1510.06112

zbMATH Keywords

binary dataexponential familyprincipal component analysis (PCA)logistic PCA

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Applications of statistics to biology and medical sciences; meta analysis (62P10) Generalized linear models (logistic models) (62J12)

Cites Work

• A feasible method for optimization with orthogonality constraints
• Generalized low rank models
• Principal component analysis.
• Title not available (Why is that?)
• On Consistency and Sparsity for Principal Components Analysis in High Dimensions
• Optimization
• Probabilistic Principal Component Analysis
• Principal component analysis of binary data by iterated singular value decomposition
• Title not available (Why is that?)
• Sparse logistic principal components analysis for binary data

Cited In (6)

• Bayesian mean-parameterized nonnegative binary matrix factorization
• Dimension reduction for longitudinal multivariate data by optimizing class separation of projected latent Markov models
• Safety signal detection with control of latent factors
• Percolate: an exponential family JIVE model to design DNA-based predictors of drug response
• Maximum A Posteriori Inference of Random Dot Product Graphs via Conic Programming
• Probability-enhanced sufficient dimension reduction for binary classification

Uses Software

• R
• LowRankModels
• Pyglrm

   Recommendations

• Probability-enhanced sufficient dimension reduction for binary classification 👍 👎
• Explicit dimension reduction and its applications 👍 👎
• Dimensionality reduction and topographic mapping of binary tensors 👍 👎
• Methods for Binary Multidimensional Scaling 👍 👎
• Title not available (Why is that?) 👍 👎
• A nonlinear approach to dimension reduction 👍 👎
• Title not available (Why is that?) 👍 👎

This page was built for publication: Dimensionality reduction for binary data through the projection of natural parameters

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2201560)