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: declaring a person period dataset and fitting discrete-time survival models with st and MI


From   Steve Samuels <sjsamuels@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: declaring a person period dataset and fitting discrete-time survival models with st and MI
Date   Thu, 5 Aug 2010 11:55:08 -0400

MI.   With data in the person-period format, --cloglog- does not take
a "time" variable. See "lesson 8" referenced below

 The _d1 _d2 dummies are used as covariates so that you can get a
separate estimate of the probability in each period.


See Chapter 9, especially Section 9.3, of Stephen Jenkins's wonderful
book "Survival Analysis" at
http://www.iser.essex.ac.uk/files/teaching/stephenj/ec968/pdfs/ec968lnotesv6.pdf

and his lesson 8 on setting up the analysis with Stata at
http://www.iser.essex.ac.uk/survival-analysis

all at:
http://www.iser.essex.ac.uk/study/resources/module-ec968

Steve

-- 
Steven Samuels
sjsamuels@gmail.com
18 Cantine's Island
Saugerties NY 12477
USA
Voice: 845-246-0774
Fax:    206-202-4783


On Thu, Aug 5, 2010 at 9:08 AM, Clara Barata
<maria_barata@mail.harvard.edu> wrote:
> Hi!
> I am trying to fit some survival analysis models to a person period
> dataset using stata/mp 11. My ultimate goal is to be able to use MI
> for data imputation, then analyze the data using mi estimate: together
> with a survival analysis command. My two questions are the following:
>
> 1)  How do i declare the data to be person-period with multiple
> dummies for time? stset seems to require the specification of just one
> time variable. Should I just use _period?
> 2) Is there a discrete-time survival model option within the st
> commands? I was using the command "dthaz" for discrete time models,
> but that does not seem compatible with "mi estimate".
>
> Thanks, Clara
>
> Here's my syntax and output for creating the person period dataset:
>
> * I used the command prsnperd to create the person-period dataset as follows
>
> . list newid time evercc censor if newid==1002 | newid==1007 |
> newid==1086 , noobs
>
>  +--------------------------------+
>  | newid   time   evercc   censor |
>  |--------------------------------|
>  |  1002      5        0        1 |
>  |  1007      4        1        0 |
>  |  1086      1        1        0 |
>
>
> prsnperd newid time censor
>
> list newid _period _Y _d1 _d2 _d3 _d4 _d5 _d6 if newid==1002 |
> newid==1007 | newid==1086 , noobs
>
> +----------------------------------------------------------+
> newid   _period   _Y   _d1   _d2   _d3   _d4   _d5   _d6
> ----------------------------------------------------------
> 1002         1    0     1     0     0     0     0     0
> 1002         2    0     0     1     0     0     0     0
> 1002         3    0     0     0     1     0     0     0
> 1002         4    0     0     0     0     1     0     0
> 1002         5    0     0     0     0     0     1     0
> ----------------------------------------------------------
> 1007         1    0     1     0     0     0     0     0
> 1007         2    0     0     1     0     0     0     0
> 1007         3    0     0     0     1     0     0     0
> 1007         4    1     0     0     0     1     0     0
> 1086         1    1     1     0     0     0     0     0
> +----------------------------------------------------------+
> *
> *   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/


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