Maximum Likelihood Estimate of Precision Matrix and Correlation Parameters for Given Network

mle(data,network,heter,ty,tole,lower,upper)

Arguments

data

Data matrix in which the first column is subject id, the second column is the time points of observation. Columns 2 to (p+2) is the observations for p variables.

heter

Binary variable TRUE or FALSE, indicating heterogeneous model or homogeneous model is fitted. In heterogeneous model, subjects are allowed to have his/her own temporal correlation parameter tau_i; while in homogeneous model, all the subjects are assumed to share the same temporal correlation parameter,i.e., tau_1=tau_2=...tau_m.

network

The structure of precision matrix

tole

Error tolerance for determination of convergence of EM algorithm

lower

Lower bound for prediction of correlation parameter tau

upper

Upper bound for prediction of correlation parameter tau

ty

Type of correlation function

Value

A list which include the maximum likelihood estimate of precision matrix, correlation parameter tau. If heter=TRUE, the output also include the estimate of alpha where tau~exp(alpha)

Author

Jie Zhou