Sounds like you want
help("tryCatch") # Catch the error and do something
help("next") # Go to the next value of the surrounding loop
Hope this helps a little.
Allan
On 31/08/10 19:34, karena wrote:
I have the following code:
-----------------------------------------------------------------------------------------------------
result<- matrix(NA, nrow=1, ncol=5)
for(i in 1:(nsnp-1)) {
for(j in (i+1):nsnp){
tempsnp1<- data.lme[,i]
tempsnp2<- data.lme[,j]
fm1<- lme(trait~sex+age+rmtemp.b+fc+tempsnp1+tempsnp2+tempsnp1*tempsnp2,
random=~1|famid, na.action=na.omit)
fm2<- lme(trait~sex+age+rmtemp.b+fc+tempsnp1+tempsnp2, random=~1|famid,
na.action=na.omit)
result[1,1]<- i
result[1,2]<- j
result[1,3]<- colnames(data.lme)[i]
result[1,4]<- colnames(data.lme)[j]
result[1,5]<- 2*(summary(fm1)$logLik-summary(fm2)$logLik)
if (i==1& j==2) results<- result
if (i!=1 | j!=2) results<- rbind(results, result)
}
-----------------------------------------------------------------------------------------------------
after submitting this code, I got the error message saying "Error in
MEEM(object, conLin, control$niterEM) :
Singularity in backsolve at level 0, block 1", I know this might be some
issue due to the missing values. what I want to do is to skip the 'bad'
variable, for example, I got this error message when j=116, so I just wanna
ignore variable116, is there a statement that can do this: when getting this
error message 'Error in MEEM...........', directly goto the next variable?
thank you,
karena
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
