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: decomposition with fixed effects model

From   Megan Sheahan <>
Subject   st: decomposition with fixed effects model
Date   Thu, 6 Feb 2014 15:02:59 -0500


I am interested to know if any Stata or user-written command allows me
to decompose r-squared when using a fixed effects regression model.

Here is the simple model I run: xtreg y x1 x2 x3, fe

I would like to know the contribution of each of the variables -- x1,
x2, and x3 -- in explaining the expected variance of y.

With a standard OLS regression, I've been able to use the use-written
command "rego," which calculates Shapley values.

A simple model like this: reg y x1 x2 x3

Can be decomposed like this: rego y x1 \ x2 \ x3

This command does not seem compatible with fixed effects estimation.
Any suggestions for doing something similar with a fixed effects model
would be most useful.

Many thanks in advance,
*   For searches and help try:

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