Bookmark and Share

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

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

st: conditional SE of y|X in glm

From   Marco Ventura <>
Subject   st: conditional SE of y|X in glm
Date   Tue, 24 Apr 2012 10:31:19 +0200

Dear Stata users,
I'm having the following problem:
from a GLM estimate I want to retrieve the conditional standard error of y given the covariates. If I do

gen sigma=sqrt(e(dispers))

do I always get the right thing independently of any family and link?
Should I correct it by sqrt(e(dispers)* (_N-1)/_N)?
And do you think I should instead use the Pearson residuals such as

gen sigma=sqrt(e(dispers_p))

Thank you very much of your courtesy.
Kind Regards,

Marco Ventura
Researcher at ISTAT
Italian National Institute of Statistics
Econometric Studies and Economic Forecasting Division
P.zza Dell'Indipendenza 4, 00185 Rome, Italy
tel: (+39)064673 3665
fax: (+39)06233232 419

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index