[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Cluster option in Regress command
----- Original Message -----
From: "Mark Schaffer" <M.E.Schaffer@hw.ac.uk>
Sent: Monday, June 09, 2003 5:46 PM
Subject: Re: st: Cluster option in Regress command
> Actually, either approach is valid (depending on the assumptions one
> makes), and in principle can even be used at the same time. See, e.g.,
> Arellano in the Oxford Bulletin of Economics and Statistics, 49:4 (1987),
> who proposed using the cluster-robust approach to the fixed effects
> estimator. A more recent source is Wooldridge, Econometric Analysis of
> Cross Section and Panel Data (2002), p. 193.
> Curiously, as of version 7, Stata's xtreg didn't have a cluster option. I
> don't know if this is true in version 8.
-xtreg- in version 8 does not have cluster option. However, -areg- does
allow fixed effects and clustered observations.
* For searches and help try: