Statalist


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

RE: st: Interpretation of regressionmodel of ln-transformed variable


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Interpretation of regressionmodel of ln-transformed variable
Date   Wed, 5 Nov 2008 19:26:05 -0000

I doubt that anything is wrong with Tony's model except that
-eform("exp(b)")- should just be -eform-.

Nick

roland andersson

Peter and Maarten

I am sorry Peter. Your model is not accepted by Stata. I tried
different alternativ without success.

However I tried Maarten GLM model

xi: glm studytime i.drug c_age cons, family(gaussian) link(log) nocons
eform

on my data and got a different result compared to the regress of the
lnLOS. Now laparoscopy has shorter LOS. Which method is correct?
						

2008/11/5 Lachenbruch, Peter <Peter.Lachenbruch@oregonstate.edu>:

> The issue seems to be that hospitals have a closure date on stay when
> you are doing a study after patients are certain (or almost certain)
to
> have been discharged (e.g., all records are from admissions at least a
> year old).
>
> An alternative model might fit the reciprocal of the mean rather than
> the log of the observations (thus obviating problems with 0 days of
stay
> - e.g. an outpatient visit to the ER)  in this case you could use
> generalized linear models to get
> xi: glm LOS  lapscopic i.appdgn age agesq cons, eform("exp(b)")
> link(power -1) nocons
>
>
> Tony

*
*   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