Keep Coding

I saw many messages regarding the message error "arguments imply differing number of rows" but here it's slightly different. I would like to maximize a likelihood function, which includes cumulative products over subgroups. However at the end I get the message "arguments imply differing number of rows: 2, 413". I have 413 lines in my dataset. This message also comes when I do it with some random variables from my dataset.
Here is the code :
fcontriloglik = function(para){
S_t<<-NA ## I try to create a global variable
Define parameters :
betarisk1 = para[1:2]
sigmarisk1 = para[3:21]
Treat1 = para[22]
Compute parts of the likelihood :
Xb1 = x%*%betarisk1
Xsig1 = sigma%*%sigmarisk1
XbT1 = treatment*Treat1
exbTreat1 = exp(Xb1 + Xsig1 + XbT1)
Contributions :
h1 = 1 - exp(-exbTreat1)
S_t <<- (1 - h1)
I want to make a cumulative product over individuals :
datatest<-ddply(newdata, .(id), transform, cumSt=cumprod(S_t))
cumulativesum<-data.frame(datatest$cumSt)
Compute the likelihood :
l1 <- (h1/(1-h1))^risk * cumulativesum
loglik = log(l1)
return(l1)
}
Return the value of the likelihood with initial values I set at the beginning :
fcontriloglik(stval)
The error comes from "cumSt=cumprod(S_t)" which it is not able to compute.
Thanks a lot !