Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: ivreg2

From   "Steve Stillman" <>
To   <>
Subject   st: RE: ivreg2
Date   Fri, 7 Jul 2006 02:37:56 +1200

Hi Valerio.  

As a co-author of ivreg2, I'll apologise for the slight ambiguity in the help-file.  ivreg2 does not estimate fixed effects or random effects panel data models.  it is able to produce standard errors that account for autocorrelation within panels hence the line in the help file.  

Unfortunately, the advance features you are looking for do not mesh well with the random effects estimator, hence the reason why xtivreg2 does not deliver in this area.  My co-author, Mark Schaffer, has spent some time trying to code up these features for random effects models, but I do not know how much progress he has made in this area.  He is active on statalist, so will likely respond to this email once he gets the chance.


-----Original Message-----
[]On Behalf Of valerio
Sent: Friday, July 07, 2006 1:06 AM
Subject: st: ivreg2

question regarding ivreg2:

STATA help says "ivreg2 may be used with time series or panel data, in which 
case, the data must be tsset before using ivreg2"

it means that ivreg2 can handle panel data, but: which kind of panel estimator 
ivreg2 will deliver??? I assumed that it does a random effect est., indeed, if 
i do xtivreg,re on the same data i obtain the same coefficients. Clearly, i 
would prefer to use ivreg 2, instead of xtivreg,re, since with the first method 
i can utilize a lot of options regarding residuals. 

NOTE: i can't use xtivreg2 since xtivreg2 does not deliver re estimator, just 
FE or FD!!!

thank you


*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index