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, is already up and running.

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

Re: st: Modeling % data

From   Austin Nichols <>
Subject   Re: st: Modeling % data
Date   Wed, 22 Sep 2010 11:17:56 -0400

Marlis Gonzalez Fernandez <>:

Quantile regression via -qreg- makes sense, though then you are
modeling how conditional quantiles change as a fn of predictors, not
the conditional mean.  If you want to model the conditional mean,
consider a GLM (-help glm-), which offers one way to model fractional
outcomes via the fractional logit; see e.g.
and see also -locpr- on SSC for an alternative conditional mean model
as a fn of one predictor.

On Wed, Sep 22, 2010 at 11:03 AM, Marlis Gonzalez Fernandez
<> wrote:
> My outcome variable is a % (% error in a language test).  We do have many 0 and 100.  I need to be able to do a multiple regression to adjust for known predictors of the variable vs. the predictors of interest.
> It was suggested that I use qreg.  I've done so and it seems to work.  Any thoughts?  I can provide more details if necessary.

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index