Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: How to find the best transformation for each variable in 120 periods


From   "JVerkuilen (Gmail)" <jvverkuilen@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: How to find the best transformation for each variable in 120 periods
Date   Fri, 22 Feb 2013 10:27:00 -0500

On Fri, Feb 22, 2013 at 9:44 AM, Xixi Lin <winnielxx@gmail.com> wrote:
> Hi,
>
> I am trying to find the best transformation for each variable in 120
> periods. I know that command ladder gives the following result:

Transforming the dependent variable before you have a notion of what
the regression model will be and the deviations from it is generally a
useless endeavor. A good bit of the non-Gaussian structure in data can
be due to the regressors and once you regress on them the data may be
much better behaved. The key assumption is Gaussian errors, not
Gaussian dependent variables unconditionally. (And even that
assumption isn't *that* important, just so long as it's not violated
too badly.)

If you're planning on using transformations, I highly recommend
getting a copy of A. C. Atkinson's book Plots, Transformations and
Regression: An Introduction to Graphical Methods of Diagnostic
Regression Analysis (1986, Oxford University Press).
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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