A0392
Title: Statistical analysis for a penalized EM algorithm in high dimensional mixture linear regression model
Authors: Ning Wang - Beijing Normal University (China) [presenting]
Abstract: The expectation-maximization (EM) algorithm and its variants are widely used in statistics. In high-dimensional mixture linear regression, the model is assumed to be a finite mixture of linear regression forms, and the number of predictors is much larger than the sample size. The standard EM algorithm, which attempts to find the maximum likelihood estimator, becomes infeasible. A penalized EM algorithm is devised, and its statistical properties are studied. Existing theoretical results of regularized EM algorithms often rely on dividing the sample into many independent batches and employing a fresh batch of samples in each iteration of the algorithm. The algorithm and theoretical analysis do not require sample-splitting. The method and theory are also extended to multivariate response cases. The proposed methods also have encouraging performances in numerical studies.