Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: OLS assumptions not met: transformation, gls, or glm as solutions?


From   "JVerkuilen (Gmail)" <[email protected]>
To   [email protected]
Subject   Re: st: OLS assumptions not met: transformation, gls, or glm as solutions?
Date   Mon, 17 Dec 2012 10:10:37 -0500

-robust- won't change the predicted values or residuals at all just
fix up the standard errors, but if the regression itself is not
sensible then these are

I think you'll probably have to lose OLS and switch to a different
model, such as a gamma GLM, or else transform. Either way you need to
think carefully about what you want to do.

Oh, one transform I'll mention that not widely used is the inverse
hyperbolic sine, which is defined for zero and negative values. It
behaves like the square root near 0 and like the log far away from 0.
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index