Statalist The Stata Listserver


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

Re: st: stset with multiple spells


From   "Chris Chung" <cchungc403@rogers.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: stset with multiple spells
Date   Tue, 25 Apr 2006 15:46:53 -0400

You don't want to reset the second risk set as 0 as you need to incorporate the history of the individuals that experienced the first event. Typically, event history models analyze the occurrence and timing of an event that may be experienced only once; the observed survival or failure times are assumed to be independent. However, in the multiple failure time data, observed failure times within the same individual cannot be assumed to be independent. Note that repeated event data are a special type of event history data, which records the timing of transitions between two or more discrete states. If you reset the second risk set as 0, you arbitrarily treat the individuals that experienced the first event and others which did not experience it (or have yet to experience it) the same. So, you better code them with sequential event numbers, rather than resetting to 0.



Hope it helps,

Chris Changwha Chung





----- Original Message ----- From: "Juan Manuel Jauregui" <juan.manuel.jauregui.2006@anderson.ucla.edu>
To: <statalist@hsphsun2.harvard.edu>
Sent: Tuesday, April 25, 2006 2:59 PM
Subject: Re: st: stset with multiple spells



Thank you so much, Chris. The paper you suggest is great. It's very long, so I
couldn't read it in detail, but seems that it's going to help me a lot,
especially considering that it brings the codes. However, at least at first
sight, it doesn't seem to help me with the issue of setting the analysis time
to zero again when the individual enters the second period of time at risk. I
mean, after the first failure, and a period of time of not being at risk. Do
you know how can I do it?

Thanks a lot.

Juan




Quoting Chris Chung <cchungc403@rogers.com>:


Cleves' STB-49 should be a good place to start. You may also want to look
at:
Ezell, M.E., Land, K.C., & Cohen, L.E. 2003. Modeling multiple failure time
data: A Survey of variance-corrected proportional hazards models with
empirical applications to arrest data. Sociological Methodology, 33:
111-167.

HTH,
Chris


----- Original Message -----
From: "Sara Mottram" <s.mottram@cphc.keele.ac.uk>
To: <statalist@hsphsun2.harvard.edu>
Sent: Monday, April 24, 2006 2:57 AM
Subject: Re: st: stset with multiple spells


> Hi Juan,
>
> I think that I am trying to do a similar analysis where I look at the
> number of consultations with a doctor for a certain condition, over a
> three year period. You do not say which version of Stata you are using > but
> I think this type of analysis is possible in Version 9 by considering > the
> data as multiple failures. However, I haven't got any further than this
> myself. I believe that exactly which command and options you need to > use
> depends on the precise question. You might also want to look at the > paper
> by Jos Twisk. "Twisk J, Smidt N, de Vente W. Applied analysis of > recurrent
> events: a practical overview. J. Epidemiol. Community Health 2005;
> 59;706-10". which can be found at
> http://jech.bmjjournals.com/cgi/reprint/59/8/706.
> Hope this helps and perhaps someone else could clarify the situation
> further.
>
> Sara
>
>
> Juan Manuel Jauregui wrote:
>> My fellow Stata lovers,
>>
>> I'm trying to stset my dataset, and I'm having trouble. I already read
>> the [st]
>> manual, the great book by Cleves et al (2004) and also his excelent
>> review of
>> this issue at the FAQ (reproducing the STB-49) and even a search of >> the
>> Statalist history and other websites wasn't fruitful.
>>
>> Here is my problem:
>>
>> I have an unbalanced panel of almost 200 individuals (countries in >> this
>> case)
>> with 100 records each (25 years of quarterly data) and I want to study
>> the
>> duration of some periods of financial distress. I have an event that
>> marks the
>> begining of each crisis, and another event that marks its end. This
>> crises
>> periods appear more than once for each country and they are separated
>> from each
>> other by periods of tranquility, that is, analysis time should begin >> from
>> 0 when
>> a new crisis begins.
>>
>> The closest thing I think I can do is to take each episode as an
>> individual and
>> take countries as groups and use a shared frialty model, but that
>> wouldn't be
>> exactly what I want to do. I'd prefer to take it as repeated episodes >> of
>> the
>> same individual.
>>
>> Any help will be greatly appreciated.
>>
>> Thanks a lot!
>>
>> Juan Manuel Jauregui
>> Doctoral Student
>> Global Economics and Management
>> UCLA Anderson School of Management
>> *
>> * 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/
>>
>
> --
> Sara Mottram Research Assistant: Biostatistics
> Primary Care Musculoskeletal Research Centre
> Primary Care Sciences
> Keele University
> Staffordshire, ST5 5BG
> Tel: 01782 584711
> Fax: 01782 583911
> *
> * 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/

*
* 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/


*
* 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/
*
*   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