Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

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

st: Which estimator to use for correlation among panels?

From   "Klaus Giesler" <>
Subject   st: Which estimator to use for correlation among panels?
Date   Mon, 10 Oct 2011 15:22:25 +0200

Hi everyone,

I got a balanced panel. Its time series dimension spans over t, t = 1, ...,
T, the cross-section spans over i, i = 1, ..., I.
The dependent variable contains the exact same time series in every panel.
Therefore, it does not vary with i, say, it's constant over i.

I suspect that there is autocorrelation within the groups of an explanatory
variable. Here's my (reduced) model:

y_it = b0 + b1 * x_it + u_it,

where y_it = y_jt for all i, j. Then, I suspect cor(x_it, x_jt) != 0 for at
least some i, j. My question is: What estimator do I use in Stata for that?
I need an estimator that controls for this possibly present correlation but
still returns "interpretable" coefficients. I played around with xtpcse,
xtdpdsys and xtdpd, but I'm not sure if they are really suited for my
purpose. Is there anyone that could point this out for me?


NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie!		
Jetzt informieren:
*   For searches and help try:

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