Premium
Sparse estimation of multivariate Poisson log‐normal models from count data
Author(s) -
Wu Hao,
Deng Xinwei,
Ramakrishnan Naren
Publication year - 2018
Publication title -
statistical analysis and data mining: the asa data science journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.381
H-Index - 33
eISSN - 1932-1872
pISSN - 1932-1864
DOI - 10.1002/sam.11370
Subject(s) - count data , multivariate statistics , univariate , statistics , poisson regression , multivariate normal distribution , poisson distribution , latent variable , mathematics , computer science , multivariate analysis , covariance , population , demography , sociology
Modeling data with multivariate count responses is a challenging problem because of the discrete nature of the responses. Existing methods for univariate count response cannot be easily extended to the multivariate case since the dependence among multiple responses needs to be properly accommodated. In this paper, we propose a multivariate Poisson log‐normal regression model for multivariate count responses by using latent variables. By simultaneously estimating the regression coefficients and inverse covariance matrix over the latent variables with an efficient Monte Carlo EM algorithm, the proposed model takes advantage of the association among multiple count responses to improve the model prediction accuracy. Simulation studies and applications to real‐world data are conducted to systematically evaluate the performance of the proposed method in comparison with conventional methods.