On 9 February 2011 18:33, Philipp Grueber <philipp.grue...@ebs.edu> wrote: > thanks for the quick answer & sorry for the mistake. Please find a corrected > version of the code below. Unfortunately, the model still does not work – > due to an error I believed to have overcome: “In log(2 * pi * sig2[i]) : > NaNs produced”
You should make sure that "2 * pi * sig2" is always positive, e.g. by using a different parametrisation for sig2, e.g. sig2[i] <- exp( omega + alpha*res[i-1]^2 + beta*sig2[i-1] ) > ... and how can I avoid the for-loop? > for (i in 2:99) { > sig2[i] <- omega + alpha*res[i-1]^2 + beta*sig2[i-1] > ll[i] <- -1/2*log(2*pi*sig2[i]) - 1/2*res[i]^2/sig2[i] > } I have no idea for sig2 but you could move ll out of the loop: > ll[2:99] <- -1/2*log(2*pi*sig2[2:99]) - 1/2*res[2:99]^2/sig2[2:99] /Arne -- Arne Henningsen http://www.arne-henningsen.name ______________________________________________ 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.