Statalist The Stata Listserver

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

st: RE: xtreg, re with cluster option

From   "Feiveson, Alan H. (JSC-SK311)" <>
To   <>
Subject   st: RE: xtreg, re with cluster option
Date   Tue, 25 Jul 2006 12:59:02 -0500

Elena - 

If the dependent variable is one observation for each individual, I
don't believe a panel model applies, so there would be no need of
cluster( ). It doesn't matter what the heirarchy is for the explanatory
variables. With homoscedastic errors and a normal distribution
assumption you can just use the ordinary least squares  command -reg-.
On the other hand if your dependent variable has a multilevel structure
then you could use -xtreg-(2 levels, random intercept only) or more
generally -xtmixed- where you can have different random effect variances
and random slopes at each heirarchical level. 

Al Feiveson

-----Original Message-----
[] On Behalf Of Elena Giarda
Sent: Tuesday, July 25, 2006 3:31 AM
Subject: st: xtreg, re with cluster option

Dear Statalisters,

I have to run a panel data regression with random effects of individual
wages on a set of explicative variables which include a macro variable
(the unemployment rate).

Note that I use Stata8.

I'm facing the problem of the different level of aggregation of
individual wages and aggregate unemployment rate. With a pooled
regression I'm aware that I can use the "cluster" option: e.g.
cluster(year). Unfortunately this option is not availble for xtreg in

Has anyone ever faced this problem? Do you have any suggestions on how
to proceed?

Many thanks!
Elena Giarda

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index