[R] Freeman-Tukey arcsine transformation

[Inman, Brant A. M.D.]

The point of the given transformation is not so much for normality as it
is for variance stabilization.  The variance of the Freeman-Tukey
transform depends only on the denominator of the proportion in
question...something that can be used to advantage. 


Brant

-----Original Message-----
From: Peter Dalgaard [mailto:p.dalgaard at biostat.ku.dk] 
Sent: Tuesday, March 13, 2007 3:36 PM
To: Bos, Roger
Cc: Inman, Brant A. M.D.; r-help at stat.math.ethz.ch
Subject: Re: [R] Freeman-Tukey arcsine transformation

Peter Dalgaard wrote:
> Bos, Roger wrote:
>   
>> I'm curious what this transformation does, but I am not curious
enough to pay $14 to find out.  Someone once told me that the arcsine
was a good way to transform data and make it more 'normal'.  I am
wondering if this is an improved method.  Anyone know of a free
reference?
>>
>>  
>>   
>>     
> Well, a paper copy of the American Statistician (1978) would be free
in 
> some sense....
>
> In the meantime I got detached from JSTOR (i.e., I went home), and I'm

> not prepared to jump through the relevant hoops for remote access at 
> this point, but AFAIR it was a relatively trivial version of the
simple 
> arcsine transform, something like replacing asin(r/n) with the average

> of asin(r/(n+1)) and asin((r+1)/(n+1)). The point of the paper is that

> you can invert explicitly for r/n if n is known.
>
>   
Well, except for a couple of sqrt() it seems....
>>             /The American Statistician/, Vol. 32, No. 4. (Nov.,
1978),
>>             p. 138. 
>>
>>
>>             Stable URL:
>>
http://links.jstor.org/sici?sici=0003-1305%28197811%2932%3A4%3C138%3ATIO
TFD%3E2.0.CO%3B2-Z
>>
>>
>>     
>
> ______________________________________________
> R-help at stat.math.ethz.ch 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.
>