robustETM-package {robustETM}

Robust Exponential Tilt Mixture Model

Description

The package robustETM consists of the functions to perform pseudolikelihood based EM test for homogeneity in generalized exponential tilt mixture models.

Details

Package:	robustETM
Type:	robustETM
Version:	1.0
Date:	2016-03-27
License: GPL>=2

Testing for homogeneity in generalized exponential titl mixture model

Motivated by analyses of DNA methylation data, we propose a semiparametric mixture model, namely the generalized exponential tilt mixture model, to account for heterogeneity between differentially methylated and non-differentially methylated subjects in the cancer group, and capture the differences in higher order moments (e.g. mean and variance) between subjects in cancer and normal groups. A pairwise pseudolikelihood is constructed to eliminate the unknown nuisance function. To circumvent boundary and non-identifiability problems as in parametric mixture models, we modify the pseudolikelihood by adding a penalty function. In addition, test with simple asymptotic distribution has computational advantages over permutational test for high-dimensional genetic and epigenetic data. We propose a pseudolikelihood based expectation-maximization test, and show the proposed test follows a simple chi-squared limiting distribution.

The methods contains in function sim are:

• The proposed PLEMT test (pseudolikelihood based EM test)

• The t-test

• The modified empirical likelihood ratio test

• The empirical likelihood ratio test

• The logistic regression test

• The Wilcoxon test

• The F test

• The KS test

Author(s)

Chuan Hong
Yong Chen
Yang Ning
Hao Wu Maintainer: Chuan Hong <[e-mail]>

References

Hong, C., Chen Y., Ning Y., Wang S., Wu H. and Carroll R.J. (2016). PLEMT: A novel pseudolikelihood based EM test for homogeneity in generalized exponential tilt mixture model (in preparation).