[R] Problem about Box-Cox transformation (topic in html form)

[Dieter Menne]

Saji Ren wrote:
> 
> 
> Recently, I want to perform a transformation on my data to make it more
> normal, meanwhile the order statistics is unchanged. So I decided to use a
> box-cox transformation.
> ...
> Then I compared the result with original data, and it really confused me:
>  http://n4.nabble.com/file/n1011015/start%2Bvalue%2Bproblem.jpeg 
> The left is my original data sample, you can see that it is symetric and
> the mean is close to 0. It just that the spread is large (there are
> outliers).
> The right is the transformed data, and the distribution is obviously no
> normal.
> 
> 

This has nothing to do with box.cox.powers.

summary(rnorm(100,0,200)^0.95)

gives about 50 NA from the negative numbers which are not plotted in hist(),
and only the positive are plotted, so you get the skewed distribution.

As an aside, a transformation of 0.95 is rarely worth the pain of having to
explain it to your audience later, even if it is "significant".

Try 

library(fortunes)
fortune(234) # Hi Peter, fortunes needs a "search"


Please do not use HTML mail.

Dieter


-- 
View this message in context: http://n4.nabble.com/Problem-about-Box-Cox-transformation-topic-in-html-form-tp1011015p1011117.html
Sent from the R help mailing list archive at Nabble.com.