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: Quantile regression


From   "JVerkuilen (Gmail)" <[email protected]>
To   [email protected]
Subject   Re: st: Quantile regression
Date   Sat, 22 Sep 2012 10:40:26 -0400

On Sat, Sep 22, 2012 at 10:33 AM, Nick Cox <[email protected]> wrote:
> The choice need not seem so stark. With a link function approach, as
> with -poisson- or -glm, link(log)-, you can fit on a log scale and
> report on the original scale. I would guess that this would make most
> sense to clinicians and patients alike in this example.

Ah I certainly didn't mean to make it look that way. Yes a gamma GLM
with log link is also a very good choice. What I like about quantile
regression is that as long as it's assumptions are reasonably met,
it's relatively immune to choices of the exact distribution. But it
can be tricky to interpret.



-- 
JVVerkuilen, PhD
[email protected]

"Out beyond ideas of wrong-doing and right-doing there is a field.
I'll meet you there. When the soul lies down in that grass the world
is too full to talk about." ---Rumi
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index