lrt_stat: Calculate the likelihood ratio test statistic.

In tensr: Covariance Inference and Decompositions for Tensor Datasets

Description

Calulate the likelihood ratio test statistic for Kronecker structured covariance models.

Usage

lrt_stat(sig_null, sig_alt, p)

Arguments

sig_null	A numeric. The MLE of the total variation parameter under the null (the standard deviation version).
sig_alt	A numeric. The MLE of the total variation parameter under the alternative (the standard deviation version).
p	A vector of integers. The dimension of the array.

Details

The LRT statistic is the exact same for all elliptically distributed Kronecker structured covariance models (not just the normal). The distribution of the likelihood ratio test statistic does change.

Value

A numeric. The likelihood ratio test statistic.

Author(s)

David Gerard.

References

Gerard, D., & Hoff, P. (2016). A higher-order LQ decomposition for separable covariance models. Linear Algebra and its Applications, 505, 57-84. https://doi.org/10.1016/j.laa.2016.04.033 http://arxiv.org/pdf/1410.1094v1.pdf

See Also

holq for obtaining the MLE of the total variation parameter.

lrt_null_dist_dim_same for getting the null distribution of the likelihood ratio test statistic.

tensr documentation built on May 2, 2019, 2:32 p.m.