Hi All, I have a dataset consisting of abundance counts of a fish and I want to test if my data are poisson in distribution or normal.
My first question is whether it is more appropriate to model my data according to a poisson distribution (if my test says it conforms) or use transformed data to normalise the data distribution? I have been using the vcd package gf<-goodfit(Y,type= "poisson",method= "MinChisq") but i get the following error message Warning message: In optimize(chi2, range(count)) : NA/Inf replaced by maximum positive value I then binned my count data to see if that might help V1 V2 1 5 34 2 10 30 3 15 10 4 20 8 5 25 7 6 30 0 7 35 3 8 40 2 9 45 3 10 50 1 11 55 0 12 60 1 but still received an error message Goodness-of-fit test for poisson distribution X^2 df P(> X^2) Pearson 2573372 33 0 Warning message: In summary.goodfit(gf) : Chi-squared approximation may be incorrect Am I getting caught out because of zero counts or frequencies in my data? Andy -- Andrew Halford Ph.D Associate Research Scientist Marine Laboratory University of Guam Ph: +1 671 734 2948 [[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.