Skew as they are, your data certainly don't look normal. Try lognormal.
The chi-square test gives good results when all counts are 5 or more,
hence the warning.
At 12:25 AM 11/12/2010, Andrew Halford wrote:>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 at 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.
===============================================================Robert A.
LaBudde, PhD, PAS, Dpl. ACAFS e-mail: ral at lcfltd.com
Least Cost Formulations, Ltd. URL: http://lcfltd.com/
824 Timberlake Drive Tel: 757-467-0954
Virginia Beach, VA 23464-3239 Fax: 757-467-2947
"Vere scire est per causas scire"