Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Cox proportional hazard model


From   Michael McCulloch <mm@pinestreetfoundation.org>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Cox proportional hazard model
Date   Thu, 12 Aug 2010 15:34:26 -0700

Nyasha, Maarten explained very well the nature of what data you'd need to to time-to-employment analysis, which it turns out you can't measure in the context of your current project. This means you can't do survival analysis like the Cox method.

Another alternative analysis strategy is to instead study the likelihood of a person getting a job within some pre-determined time frame, using logistic regression. The dependent variable then is getting employment (yes/no), and the predictor variables are things like
		food aid
		use of AIDS medication
		and other characteristics of the person or household
Michael


On Aug 12, 2010, at 3:06 PM, Nyasha Tirivayi wrote:

Dear Maarten

Thanks for clarifying. Now I understand. I had this wrong. In that
case I think I need to ask for ideas for alternative method since I do
not have data on employment/unemployment spells.

I have realized that I have 2 period panel data on employment status.
So I have employment1 and employment2. I am interested in seeing the
effects of AIDS medication  on transitions from one state to another.

I also have another treatment variable- food aid. So you have some
households where a patient is on AIDS medication and household
receives food assistance. If I choose instead to look at the effect of
food assistance ( meaning all households in analysis will now have a
patient on AIDS medication), what kind of transition analysis can I
do.. or maybe I should simply carry out multiple regression analysis
or propensity score matching?

Kindly advise

Regards

Nyasha Tirivayi
Maastricht University

On Thu, Aug 12, 2010 at 11:56 PM, Maarten buis <maartenbuis@yahoo.co.uk > wrote:
--- On Thu, 12/8/10, Nyasha Tirivayi wrote:
So the duration of the AIDS medication taken by the patient
is the time variable, and getting employed/returning to work
is the event

The time variable and the event are not two separate concepts,
the duration is the time till the event occurs. When you look
at time till employment, then starting AIDS medication is not
the beginning of a spell, leaving the labor market, leaving
school, becoming unemployed, are the beginning of the spell.
Whether or not the patient takes AIDS medication is your
explanatory variable, not your explained variable.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------




*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


Best wishes,

Michael McCulloch, LAc MPH PhD
Pine Street Foundation
124 Pine Street
San Anselmo, CA 94960-2674
tel:	415-407-1357
fax: 	206-338-2391

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index