[R] ks.test

Ben Bolker bolker at ufl.edu
Fri Nov 16 14:22:10 CET 2007




elyakhlifi mustapha wrote:
> 
> Hello,
> I want to do normality test on my data 
> I write this but I don't understand the display of the results
> 
> ks.test(data,"pnorm")
> 
> In fact I want to know if my data is a normal distribution. I have to
> check the p-value or D?
> Thanks.
> 
> 

Example:

> r = rnorm(100)
> ks.test(r,"pnorm")

        One-sample Kolmogorov-Smirnov test

data:  r 
D = 0.1248, p-value = 0.08874
alternative hypothesis: two-sided 

  Interpretation: the probability of a D value this far from 0
(i.e., abs(D)>=0.1248) is p=0.088 under the null hypothesis
that the data are normally distributed.  Hence, we would
fail to reject the null hypothesis at the "usual" alpha=0.05
level.

  You might want to consider shapiro.test instead/in addition.

  Ben Bolker

-- 
View this message in context: http://www.nabble.com/ks.test-tf4820384.html#a13792656
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list