search for: cornatur

Hi, nlme users, I am fitting a model for repeated measures, using gls(). The var-cov structure is UNSTRUCTURED in lower triangle and identical for the diagonal, for which I still have no idea how to fit using PROC MIXED. The error message is as follows: Error in "coef<-.corNatural"(*tmp*, value = log((cStNatPar + 1)/(1 - cStNatPar))) : NA/NaN/Inf in foreign function call (arg 1) I have no idea why that will happen, since the same codes work well for other similar data. The data and the codes are as follows. I am wondering whether I can get help here. Thank...

...s=6)) datamat$A<-as.factor(rep(c(1:3),each=18,times=2)) datamat$B<-as.factor(rep(c("one","two"),each=54)) Then fitting the model: fm.datamat<-lme(values~A*B,data=datamat,random=~1|subject,correlation=corSym m()) produced this error: "Error in "coef<-.corNatural"(`*tmp*`, value = log((cStNatPar + 1)/(1 - : NA/NaN/Inf in foreign function call (arg 1)" If I skip observation subject 15 (row 15 above) the computation is successfull. Is it because of the similar values ? Does it fail to estimate a general variance structure ? Is there a b...

Dear all, I am trying to define my own corStruct which is different from the classical one available in nlme. The structure of this correlation is given below. I am wondering to know how to continue with this structure by using specific functions (corMatrix, getCovariate, Initialize,...) in order to get a structure like corAR1, corSymm which will be working for my data. Thanks in advance.