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]

Re: st: Fixed effects or random effects model?

From   Muhammad Anees <>
Subject   Re: st: Fixed effects or random effects model?
Date   Sat, 7 Jan 2012 10:43:27 +0500

If you need to use mixture of fixed and random effects, population
averaged model would be the option. Rest about panel data issues are
best answered by Panel Data Econometrics (2006) of Hadi H Baltagi,
Microeconometrics, Methods and Applications by A. Colin Cameron and
Pravin K. Trivedi and also their Microeconometrics using Stata. Rest
of the issues are shortly and well described in Stata panel data

On Fri, Jan 6, 2012 at 8:43 PM, sardar ahmad <> wrote:
> Dear all,
> I am a PhD student and want to analyse Panel data. I have just started to learn about panel data analysis, so I apologise if my question is very basic.The data is for 8 years for 280 companies, I used Hausman test to decide whether to use fixed effects or random effects. The test shows that I should use fixed effects. However, I want to include a number of control variables which are constant over time e.g. industry in which a company is operating. As Stata drops such variables from the model, so I was wondering will it be justified to use random effects instead, even if the Hausman test is significant. Or are there any other alternatives?
> Many thanks in advance.
> Ahmad
> *
> *   For searches and help try:
> *
> *
> *


Muhammad Anees
Assistant Professor/Programme Coordinator
COMSATS Institute of Information Technology
Attock 43600, Pakistan

*   For searches and help try:

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