Hi all,
I have a very large number of vectors that I want first to look fast which
distribution might be considered candidate for fitting.
I made a simple loop that checks for all vector (the code below is for one
vector and being called for each vector separately). If a good fit is found
this is dumped to a txt file so to allow me later on, see results
distList<-c("norm","exp","gamma","lnorm")
for (dist in distList) {
if (gofstat(fitdist(onVector,distr=dist))$kstest =="not rejected"){
# keep it
out<-capture.output(gofstat(fitdist(onVector,distr=distr))$ks)
print(sprintf("Saving to file %s ",filename))
cat(out,file=paste(filename,".txt",sep=""),sep="\n",append=TRUE)
}
}
the major problem is that these for loops return errors (sometimes for a given
vector a specific distribution does not make sense or the vector might be zero
or containing only many times the same element) . The easiest would be in
errors and warning just move to the next element of the for loop.
Something like that
for ()
{
if error==skip to next element
else do normal stuff
}
how I can do that in R?
Regards
Alex
[[alternative HTML version deleted]]
______________________________________________
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.
