[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: ttest or xtmelogit?

From   "Nick Cox" <>
To   <>
Subject   RE: st: ttest or xtmelogit?
Date   Wed, 12 Mar 2008 18:14:12 -0000

Steven is correct. This isn't mentioned in -transint-. 

-transint- (on SSC) is a slightly unusual package. It is just a help
file written because I wanted my (geography) students to have something
better than the rather poor coverage of transformations in the books
available to them. In fact, I haven't been able to find many accounts of
transformations that were very concise, covered the really important
ideas, but were also light on the mathematics, which is of course a
contradictory desire. Anyway, it then seemed that it might be useful a
little more widely. 

Variance-stabilisation is as Steven says the motive for the angular: it
is difficult
to imagine it arising except out of an algebraic argument, which I think
goes back 
to Fisher. So, next time around, that might merit an explanation. 


Steven Samuels

Not mentioned in -transint- is the variance-stabilizing property of  
the angular transformation: it has asymptotic variance 1/4n, which is  
not a function of p (Anscombe, 1948). If the observed proportion is r/ 
n, Anscombe showed that the arcsine of [(r + 3/8)/(n + 3/4)]^.5  is  
even better at stabilizing the variance, for moderate sample size.  
The second version has variance 1/(4n + 2).

The arcsine-transformation used to be recommended because transformed  
proportions  could be analyzed via standard ANOVA programs.  I once  
found it useful in a variance components analysis. The 'error'  
variance was a mixture of a between-sample and within sample  
(binomial) variance. With the arcsine transformation, I could  
subtract out the part attributable to binomial variation.


FJ Anscombe 1948. The transformation of Poisson, Binomial, and  
negative-binomial data. Biometrika 35:246-254
On Mar 10, 2008, at 6:02 PM, Nick Cox wrote:

> By arcsin I guess you mean the angular transformation (arcsine of  
> square
> root).
> Its use seems to have faded dramatically in recent years.
> Tukey showed that this is very close to p^0.41 - (1 - p)^0.41. That
> makes it weaker
> than the logit. My guess is that it would be an unusual dataset in  
> which
> the angular
> was much better than leaving data as is and also much better than the
> logit. It could happen,
> but it seems to be rare.
> The Tukey reference is given in -transint- from SSC.
> Nick
> David Airey
> Maybe I should not have said it was pilot data! I won't disagree, but
> when cluster number is too small (< 20) to invoke xtgee or xtmelogit
> on the observed yes/no data, or glm on the summary statistics with
> binomial family and logit link, what do you do? It seems to me there
> is a sample size between 10 and 30 clusters of yes/no data that may be
> better suited to some of the older approaches like arcsin transformed
> proportions and then ttest or ANOVA/regress. I guess that was my
> question.

*   For searches and help try:

© Copyright 1996–2019 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index