Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Reconcile Log Transformed with Untransformed Results


From   Erasmo Giambona <e.giambona@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Reconcile Log Transformed with Untransformed Results
Date   Sun, 14 Feb 2010 14:45:54 +0100

Thanks Maarten. I ran the regressions with both RHS and LHS
untransformed using both OLS and GLM with link(log). With the OLS the
coeff on X is 0.006 while with the GLM the coefficient is 0.700. I
find a bit hard to intepret the GLM coefficient. Maybe I
mistintepreted what you meant?

If you have any other hints, I would appreciate it.
Erasmo

On Sun, Feb 14, 2010 at 1:00 PM, Maarten buis <maartenbuis@yahoo.co.uk> wrote:
> --- On Sat, 13/2/10, Erasmo Giambona wrote:
>> Assuming I want to use the log specification, can I still
>> compute economic significance?
>>
>> That is, in the example, could I say that a 1 IQR change in
>> X (i.e., 0.85) is approximately associated with a 0.008*
>> 0.85=0.0068 change in Y?
>
> No, the problem is that by first transforming the dependent
> variable you are no longer modeling the average in y but the
> average in log(y). If you want to study your effects in terms
> of the original unit of y you'll have to use -glm- with the
> -link(log) option rather than use -regress- with the
> transformed dependent variable.
>
> Hope this helps,
> Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
> http://www.maartenbuis.nl
> --------------------------
>
>
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index