[R] Kronecker product in PROC MIXED and lme

Giovanni Parrinello parrinel at med.unibs.it
Mon Aug 6 09:48:58 CEST 2007

Dear All,
I am trying to translate some code in SAS about longitudinal data in R.
I have found among the predefined covariance structures for repeated 
measures in the PROC MIXED the following:
type=UN at AR(1).
'Kronecker product structure of the variance–covariance matrixi =dimni 
(E⊗V),V, where
 and V, respectively, are (q×q) and (p×p) positive denite matrices. 
The matrix  represents the variance–covariance matrix between the 
on all response variables at a given time point. It is assumed that it 
does not depend
on a particular time point and is the same for all time points. The 
matrix V is the correlation
matrix of the repeated measures on a given response variable, and is 
assumed to be the same
for all response variables.'
Is it possible?

dr. Giovanni Parrinello
External Lecturer
Medical Statistics Unit
Department of Biomedical Sciences
Viale Europa, 11 - 25123 Brescia Italy
Tel: +390303717528
Fax: +390303717488
email: parrinel at med.unibs.it

More information about the R-help mailing list