A Pseudo-EM Algorithm for Clustering Incomplete Longitudinal Data

Abstract

A method for clustering incomplete longitudinal data, and gene expression time course data in particular, is presented. Specifically, an existing method that utilizes mixtures of multivariate Gaussian distributions with modified Cholesky-decomposed covariance structure is extended to accommodate incomplete data. Parameter estimation is carried out in a fashion that is similar to an expectation-maximization algorithm. We focus on the particular application of clustering incomplete gene expression time course data. In this application, our approach gives good clustering performance when compared to the results when there is no missing data. Possible extensions of this work are also suggested.

Semiparametric Estimation and Panel Data Clustering Analysis Based on D-Vine and C-Vine

“…Shaikh et al [16] applied the clustering method based on the model to cluster the missing vertical data. They used the mixed normal distribution and improved the Cholesky method for decomposing the covariance structure and then used the EM algorithm to estimate the parameters of the method model.…”

Clustering Profiles in Generalized Linear Mixed Models Settings Using Bayesian Nonparametric Statistics

“…This method is implemented in software [89]. McNicholas [90] further extends the task of clustering longitudinal data with missing observations. Eigen-decomposition is another common type of decomposition of covariance matrix [10], [11] and obviously leads to different types of constraints of new matrices.…”