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

RE: st: Question about GLM

From   "Lee, Albert" <>
To   "" <>
Subject   RE: st: Question about GLM
Date   Tue, 17 Nov 2009 18:53:33 -0500

Many thanks.  This is very helpful.

No, I don't have anything fundamentally against delta method.  In fact, I love the concept.  I was hoping to avoid to implement it by hand, i.e., pencil and paper.  

From: [] On Behalf Of Maarten buis []
Sent: Tuesday, November 17, 2009 5:58 PM
Subject: Re: st: Question about GLM

--- On Tue, 17/11/09, Lee, Albert wrote:
> I recently estimated a fractional logit model using glm:
> Is there a way to calculate the standard error of the
> expected values without resorting to delta methods or other
> by-hand calculations?

Do you have something fundamentally against the delta method,
or do you think that it is a lot of work to implement. If the
latter is your concern, than you can consider the -predictnl-
command like in the example below:

*------------------------ begin example ----------------------
use, clear
glm safety noleft popdens, link(logit) robust
predictnl pr = invlogit(xb()), ci(lb ub)

sort popdens
twoway rarea lb ub popdens if noleft==0  || ///
       rarea lb ub popdens if noleft==1  || ///
       line pr popdens if noleft==0      || ///
       line pr popdens if noleft==1
*-------------------- end example -----------------------
( For more on how to use examples I sent to statalist see: )

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index