[R] MLE for multivariate t-distribution

christine.kohl at uni-ulm.de christine.kohl at uni-ulm.de
Fri May 28 11:51:43 CEST 2010


Hey,

I've got a problem with the estimation of a multivariate t-distribution.
I've got 200 observations vor 20 variables.
Now I want to estimate the parameters of the densityfunction of the  
multivarate t-distribution with mean=0.
I found a function mst.mle in the package sn, but here it is for a  
skwed t- distribution and the mean is also estimated.
I need a function which estimated only the degrees of freedom and the  
covarianzmatrix/ Omega.

Can anybody help me please!
Thanks a lot!



More information about the R-help mailing list