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: nonstationarity in dynamic panels

From   Svenja Gärtner <>
To   "" <>
Subject   st: nonstationarity in dynamic panels
Date   Fri, 4 Nov 2011 10:46:26 +0100

Dear Listers,

I have a panel (n=3, t=30) inlcuding the gender wage gap, economic variables (like unemployment, public spending) and dummies.
The wage gap is the dependent variable and is auto-correlated. I wanted to use a dynamic panel approach to include lags of the dependent variable and control for endogenity.
But I see the following problems:
 - most approaches are set up for large N, small T; I have small N, large T (like -xtabond-)
 - the dependent variable is non-stationary as well as some regressors, but not integrated of the same order ( so I cannot do a VEC for each cross-section)

Do you have any suggestions how to deal with such a panel?

Your help is very much appreciated.


Svenja Gärtner
PhD Student
Department of Economic History
School of Business, Economics and Law
University of Gothenburg
+46 (0)31 - 7864443
*   For searches and help try:

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