Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: duration data: stock sampling


From   "Stephen P. Jenkins" <stephenj@essex.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: duration data: stock sampling
Date   Thu, 22 May 2008 17:09:57 +0100

> -----Original Message-----
> From: Stephen P. Jenkins [mailto:stephenj@essex.ac.uk] 
> Sent: 22 May 2008 09:35
> To: 'statalist@hsphsun2.harvard.edu'
> Subject: st: duration data: stock sampling
> 
> 
> >>>>>>>>>>>>
> 
> Date: Wed, 21 May 2008 14:24:15 +0000
> From: marcel spijkerman <marcel_spijkerman@hotmail.com>
> Subject: st: duration data: stock sampling
> 
> Dear statalisters,
> 
> I have a sample of duration data on unemployment. The sample 
> shares characteristics of stock and flow sample selection. 
> That is, I have all persons that where unemployed between 1 
> january 1999 and 1 january 2007; including those who have 
> been unemployed before 1 january 1999 and still unemployed 
> between 1999 and 2007 as well as new entrants who became 
> unemployed somewhere between 1999 and 2007.
> 
> For estimation I use the specification of Jenkins for 
> discrete duration data where spells are converted to spell 
> months and estimating proceeds by estimation of binary data. 
> My question is whether for those who were already unemployed 
> before 1999, I should condition the likelihood as described 
> by Jenkins. If so, how should I do this?
> 
> <<<<<<<<<<<<<<
> 
> You have a combination of sampling types: (i) inflow sample 
> (spells begin within window of observation), and (ii) stock 
> sample with follow-up (spells starting before window of 
> observation and surviving until at least start of window). I 
> assume you know when those in sample (ii) began their spell; 
> if not, the spells are left-censored and you're stuck.  
> Otherwise, yes, the likelihood contributions for those spells 
> of type (ii) must use appropriate conditioning for "delayed 
> entry" (a.k.a. "left truncation"), as described in my website 
> materials and elsewhere.
> 
> 
> Stephen
> -------------------------------------------------------------
> Professor Stephen P. Jenkins <stephenj@essex.ac.uk>
> Director, Institute for Social and Economic Research
> University of Essex, Colchester CO4 3SQ, U.K.
> Tel: +44 1206 873374.  Fax: +44 1206 873151.
> http://www.iser.essex.ac.uk  
> Survival Analysis using Stata: 
> http://www.iser.essex.ac.uk/teaching/degree/stephenj/ec968/ 
> Downloadable papers and software: http://ideas.repec.org/e/pje7.html


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index