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: RE: BHPS

From   Yasir Zuberi <>
To   "" <>, "" <>
Subject   RE: st: RE: BHPS
Date   Tue, 1 Apr 2014 14:46:30 +0100


I hope you are well. I have used the BHPS and extracted the info sex sctype smoke nicigs pid age wave. And I have ran LPM regressions and Logit Regessions. However, I was interested in looking at the time trend by looking at rates of cessation (smoking t and not smoking at t+1) using logistic regression. is this similar to the code "xtreg cigarettes i.sctype age i.wave, i(pid) re"? Could you please give any advice in how to do this. 

In addition to this can you explain why we use random effects as apposed to fixed effects with the BHPS - when the Hausman test results indicates fixed effects. Is the reasoning because of the assumptions which are made?

Any help would be appreciated



> From:
> To:
> Subject: Re: st: RE: BHPS
> Date: Mon, 31 Mar 2014 14:01:27 +0100
> On Mon, Mar 31 2014, Yasir Zuberi wrote:
>> I was also
>> wondering if you had any resources which tell you how to add a wave
>> dummy variable into my BHPS dataset. I have variables sctype, smoker,
>> ncigs, pid,age, wave and cigarettes and now wanted to a dummy variable
>> to wave to see how it changes over time. 
> With the example I sent before, do:
> xtreg cigarettes i.sctype age i.wave, i(pid) re
> Brendan
> -- 
> Brendan Halpin, Head, Department of Sociology, University of Limerick, Ireland
> Tel: w +353-61-213147 f +353-61-202569 h +353-61-338562; Room F1-002 x 3147
> ULSociology on Facebook:
> twitter:@ULSociology
> *
> * For searches and help try:
> *
> *
> * 		 	   		  
*   For searches and help try:

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