[R] What can I use instead of ks.test for the binomial distr
(Ted Harding)
Ted.Harding at manchester.ac.uk
Sat Mar 13 22:52:37 CET 2010
Sorry -- I made a complete mess of my first suggestion below:
On 13-Mar-10 21:27:28, Ted Harding wrote:
> For testing whether x comes from a binomial distribution,
> I would suggest just using a straight chi-squared test.
> I'm not aware of a version of chisq.test() in R aimed at
> seeing whether data match a fitted binomial (or other
> specific) distribution, but it is easy to construct one;
>
> x<-rbinom(10000,10,0.5)
> phat<-sum(x)/length(x) ####!! phat will be wrong by *10!!
> dhat <- dbinom((0:10),10,0.5) ####!! Uses p=0,5 instead of phat!!
> Freq <- table(x)
> Exp <- 10000*dhat
> ChiSq <- sum(((Freq-Exp)^2)/Exp); DF <- (11-1-1)
> 1-pchisq(ChiSq,DF)
> # [1] 0.6788829
Instead:
N <- 10000
n <- 10
x<-rbinom(N,n,0.5)
phat<-sum(x)/(N*n)
dhat <- dbinom((0:n),n,phat)
Freq <- table(x)
Exp <- N*dhat
ChiSq <- sum(((Freq-Exp)^2)/Exp); DF <- (n-1-1)
1-pchisq(ChiSq,DF)
# [1] 0.8537699
(I think the rest of my previous response is OK ... ).
Ted.
[All contributions to my Hypocaffeinemia Appeal gratefull received]
--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at manchester.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 13-Mar-10 Time: 21:52:33
------------------------------ XFMail ------------------------------
More information about the R-help
mailing list