EstHyper {StructFDR}

Estimate the Hyper-parameter Using Generalized Least Squares

Description

Estimate the hyper-parameter by calling the gls function. The errors are allowed to be correlated and/or have unequal variances.

Usage

EstHyper(y, D, init.val)

Arguments

y

a vector of z-values.

D

a distance matrix defined based on the prior structure. Diagonal have to be zeros.

init.val

initial values for the transformed hyper-parameter. Default is 0.

Value

a vector of estimated hyper-parameter values plus log likelihood.

Author(s)

Jun Chen

References

Xiao, J., Cao, H., Chen, J. (2017). False discovery rate control incorporating phylogenetic tree increases detection power in microbiome-wide multiple testing. Bioinformatics, 33(18), 2873-2881.