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: Haussman test Question

From   sam sampton <>
Subject   st: Haussman test Question
Date   Thu, 16 Jun 2011 12:01:30 -0700 (PDT)

Dear Statalisters,
I have a question about the accuracy of the Haussman test when some 
time-varying, individual-unvarying variables are included in the model: 
Y_it = X_it*Beta + Z_t*gama + e_it
Z_t for instance. Can we still consider the Haussman test result valid? od can 
an over-identification test more valid: xtoverid, robust;

*   For searches and help try:

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