Sparse Bayesian infinite factor models. (English) Zbl 1215.62025

Bhattacharya, A.; Dunson, D. B.

Biometrika 98, No. 2, 291-306 (2011).

Summary: We focus on sparse modelling of high-dimensional covariance matrices using Bayesian latent factor models. We propose a multiplicative gamma process shrinkage prior on the factor loadings which allows introduction of infinitely many factors, with the loadings increasingly shrunking towards zero as the column index increases. We use our prior on a parameter-expanded loading matrix to avoid the order dependence typical in factor analysis models and develop an efficient Gibbs sampler that scales well as data dimensionality increases. The gain in efficiency is achieved by the joint conjugacy property of the proposed prior, which allows block updating of the loadings matrix. We propose an adaptive Gibbs sampler for automatically truncating the infinite loading matrix through selection of the number of important factors. Theoretical results are provided on the support of the prior and truncation approximation bounds. A fast algorithm is proposed to produce approximate Bayes estimates. Latent factor regression methods are developed for prediction and variable selection in applications with high-dimensional correlated predictors. Operating characteristics are assessed through simulation studies, and the approach is applied to predict survival times from gene expression data.

MSC:

62F15	Bayesian inference
62H25	Factor analysis and principal components; correspondence analysis
65C60	Computational problems in statistics (MSC2010)

Keywords:

adaptive Gibbs sampling; factor analysis; high-dimensional data; multiplicative gamma process; parameter expansion; regularization; shrinkage; diffuse large B-cell lymphoma