Recurrent Event Counting Process

[Asrul Akmal Shafie]

Hi,

First time posting and would be very glad for input and guidance. I am
currently analyzing my data using SPSS version 13 by Cox regression. After
consulting KLEINBAUM, D. G. & KLEIN, M. (2005) Survival analysis: a self
learning text, New York, Springer-Verlag, I have decided that the most
appropriate analysis would be recurrent event Cox regression by counting
process. The question is how do you run this analysis on the SPSS?

Thank you for your time.

Asrul Akmal Shafie
Cardiff University

[Maguin, Eugene]

Asrul,

The simple answer No. You can't run this kind of analysis directly. That
said, I wonder if spss can't be made to run it. My idea is based on the fact
that a discrete time survival analysis can be restructured to be solved
using ordinary logistic regression. At least one reference for this is Paul
Allison. It seems to me that a competing risks model might be restructured
to be solvable using either ordinal logistic regression or multinomial
logistic regression. However, I have never seen anybody do this nor have I
read about doing this. I'm not a statistician so there's lots of things I
don't know about this. If you haven't gotten a reply perhaps somebody who
has done this kind of model would reply to the list. I think that Hector had
a problem that involved this kind of analysis a while back.

All that said, you can do this sort of problem using other software,
notably, Mplus, Mlwin, perhaps, HLM or SAS. You might want to look at Judith
Singer's book, Applied longitudinal data analysis to see if she has anything
about this sort of analysis in there. There may also be free software that
can do this analysis. I'd investigate Hedeker's programs.

Good luck, Gene Maguin

[Butler, Deborah {FLNA}]

 If SPSS cannot handle it, I would suggest R.  It's a freeware package -
very powerful.  You'll need to download a specific package, but I've no
doubt it could handle it.

http://www.r-project.org/


Debbie Butler
SCP Modeling Analyst
Frito Lay Operations
972.334.3934

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Gene Maguin
Sent: Wednesday, November 01, 2006 9:43 AM
To: [hidden email]
Subject: Re: Recurrent Event Counting Process

Asrul,

The simple answer No. You can't run this kind of analysis directly. That
said, I wonder if spss can't be made to run it. My idea is based on the
fact
that a discrete time survival analysis can be restructured to be solved
using ordinary logistic regression. At least one reference for this is
Paul
Allison. It seems to me that a competing risks model might be
restructured
to be solvable using either ordinal logistic regression or multinomial
logistic regression. However, I have never seen anybody do this nor have
I
read about doing this. I'm not a statistician so there's lots of things
I
don't know about this. If you haven't gotten a reply perhaps somebody
who
has done this kind of model would reply to the list. I think that Hector
had
a problem that involved this kind of analysis a while back.

All that said, you can do this sort of problem using other software,
notably, Mplus, Mlwin, perhaps, HLM or SAS. You might want to look at
Judith
Singer's book, Applied longitudinal data analysis to see if she has
anything
about this sort of analysis in there. There may also be free software
that
can do this analysis. I'd investigate Hedeker's programs.

Good luck, Gene Maguin

[Edward Boadi]

What book will you recommend for somebody who want to learn R ?

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
Butler, Deborah {FLNA}
Sent: Wednesday, November 01, 2006 1:30 PM
To: [hidden email]
Subject: Re: Recurrent Event Counting Process


 If SPSS cannot handle it, I would suggest R.  It's a freeware package -
very powerful.  You'll need to download a specific package, but I've no
doubt it could handle it.

http://www.r-project.org/


Debbie Butler
SCP Modeling Analyst
Frito Lay Operations
972.334.3934

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Gene Maguin
Sent: Wednesday, November 01, 2006 9:43 AM
To: [hidden email]
Subject: Re: Recurrent Event Counting Process

Asrul,

The simple answer No. You can't run this kind of analysis directly. That
said, I wonder if spss can't be made to run it. My idea is based on the
fact
that a discrete time survival analysis can be restructured to be solved
using ordinary logistic regression. At least one reference for this is
Paul
Allison. It seems to me that a competing risks model might be
restructured
to be solvable using either ordinal logistic regression or multinomial
logistic regression. However, I have never seen anybody do this nor have
I
read about doing this. I'm not a statistician so there's lots of things
I
don't know about this. If you haven't gotten a reply perhaps somebody
who
has done this kind of model would reply to the list. I think that Hector
had
a problem that involved this kind of analysis a while back.

All that said, you can do this sort of problem using other software,
notably, Mplus, Mlwin, perhaps, HLM or SAS. You might want to look at
Judith
Singer's book, Applied longitudinal data analysis to see if she has
anything
about this sort of analysis in there. There may also be free software
that
can do this analysis. I'd investigate Hedeker's programs.

Good luck, Gene Maguin

[lts1]

Hi Edward,

   A few years ago I was using both S+ & R (the open source version of S+),
and I found "The Basics of S-Plus" by Krause & Olson to be useful for a
beginner. HTH.

    Best,
        Lisa

Lisa T. Stickney
Ph.D. Candidate
The Fox School of Business
     and Management
Temple University
[hidden email]



----- Original Message -----
From: "Edward Boadi" <[hidden email]>
To: <[hidden email]>
Sent: Wednesday, November 01, 2006 3:10 PM
Subject: Re: Recurrent Event Counting Process

> What book will you recommend for somebody who want to learn R ?
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
> Butler, Deborah {FLNA}
> Sent: Wednesday, November 01, 2006 1:30 PM
> To: [hidden email]
> Subject: Re: Recurrent Event Counting Process
>
>
> If SPSS cannot handle it, I would suggest R.  It's a freeware package -
> very powerful.  You'll need to download a specific package, but I've no
> doubt it could handle it.
>
> http://www.r-project.org/
>
>
> Debbie Butler
> SCP Modeling Analyst
> Frito Lay Operations
> 972.334.3934
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
> Gene Maguin
> Sent: Wednesday, November 01, 2006 9:43 AM
> To: [hidden email]
> Subject: Re: Recurrent Event Counting Process
>
> Asrul,
>
> The simple answer No. You can't run this kind of analysis directly. That
> said, I wonder if spss can't be made to run it. My idea is based on the
> fact
> that a discrete time survival analysis can be restructured to be solved
> using ordinary logistic regression. At least one reference for this is
> Paul
> Allison. It seems to me that a competing risks model might be
> restructured
> to be solvable using either ordinal logistic regression or multinomial
> logistic regression. However, I have never seen anybody do this nor have
> I
> read about doing this. I'm not a statistician so there's lots of things
> I
> don't know about this. If you haven't gotten a reply perhaps somebody
> who
> has done this kind of model would reply to the list. I think that Hector
> had
> a problem that involved this kind of analysis a while back.
>
> All that said, you can do this sort of problem using other software,
> notably, Mplus, Mlwin, perhaps, HLM or SAS. You might want to look at
> Judith
> Singer's book, Applied longitudinal data analysis to see if she has
> anything
> about this sort of analysis in there. There may also be free software
> that
> can do this analysis. I'd investigate Hedeker's programs.
>
> Good luck, Gene Maguin
>

[Butler, Deborah {FLNA}]

 It also accepts Matlab code.  You could start here in learning it:
http://www.r-project.org/  Just get in and start using it!

Just a word of caution, R uses memory only and does not write to disk,
therefore, it is NOT helpful for those of us with large datasets and
large data manipulations.  This is one of the reasons why a professional
package such as SPSS is worth the money!  I like the versatility of it,
however, and plan to use it concurrently with SPSS for special projects.
I also like that it is free. ;)


Debbie Butler
SCP Modeling Analyst
Frito Lay Operations

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Lisa Stickney
Sent: Wednesday, November 01, 2006 2:58 PM
To: [hidden email]
Subject: Re: Recurrent Event Counting Process

Hi Edward,

   A few years ago I was using both S+ & R (the open source version of
S+),
and I found "The Basics of S-Plus" by Krause & Olson to be useful for a
beginner. HTH.

    Best,
        Lisa

Lisa T. Stickney
Ph.D. Candidate
The Fox School of Business
     and Management
Temple University
[hidden email]



----- Original Message -----
From: "Edward Boadi" <[hidden email]>
To: <[hidden email]>
Sent: Wednesday, November 01, 2006 3:10 PM
Subject: Re: Recurrent Event Counting Process


> What book will you recommend for somebody who want to learn R ?
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf
Of
> Butler, Deborah {FLNA}
> Sent: Wednesday, November 01, 2006 1:30 PM
> To: [hidden email]
> Subject: Re: Recurrent Event Counting Process
>
>
> If SPSS cannot handle it, I would suggest R.  It's a freeware package
-
> very powerful.  You'll need to download a specific package, but I've
no

> doubt it could handle it.
>
> http://www.r-project.org/
>
>
> Debbie Butler
> SCP Modeling Analyst
> Frito Lay Operations
> 972.334.3934
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf

Of
> Gene Maguin
> Sent: Wednesday, November 01, 2006 9:43 AM
> To: [hidden email]
> Subject: Re: Recurrent Event Counting Process
>
> Asrul,
>
> The simple answer No. You can't run this kind of analysis directly.
That
> said, I wonder if spss can't be made to run it. My idea is based on
the
> fact
> that a discrete time survival analysis can be restructured to be
solved
> using ordinary logistic regression. At least one reference for this is
> Paul
> Allison. It seems to me that a competing risks model might be
> restructured
> to be solvable using either ordinal logistic regression or multinomial
> logistic regression. However, I have never seen anybody do this nor
have
> I
> read about doing this. I'm not a statistician so there's lots of
things
> I
> don't know about this. If you haven't gotten a reply perhaps somebody
> who
> has done this kind of model would reply to the list. I think that
Hector

> had
> a problem that involved this kind of analysis a while back.
>
> All that said, you can do this sort of problem using other software,
> notably, Mplus, Mlwin, perhaps, HLM or SAS. You might want to look at
> Judith
> Singer's book, Applied longitudinal data analysis to see if she has
> anything
> about this sort of analysis in there. There may also be free software
> that
> can do this analysis. I'd investigate Hedeker's programs.
>
> Good luck, Gene Maguin
>

[Eric Janssen]

> My idea is based on the fact that a discrete time survival analysis can be
restructured to be solved using ordinary logistic regression. At least one
reference for this is Paul Allison.

Check:
Allison P. D. (1984) "Event history analysis. Regression for longitudinal event
data", Sage University Paper #46, Series Quantitative Applications in the
social sciences.

Allison P. D. (1982) "Discrete-time methods for the analysis of event histories"
in S. Leinhardt (ed.) Sociological Methodology, San Francisco: Jossey-Bass, pp.
61-98.
A quite comprehensive introduction of the method.

Goodies:
- you can use the time IV as linear or categorical;
- you can apply multinomial logreg;
- no need to check equiprobablity assumptions as in Cox model
Baddies:
- an event may occur more than once within a single time interval - but there
are tricks to manage it properly.

Doug Massey and colleagues used discrete time event analysis in their studies of
the Mexican Migration project data base - a technique applied in Mexico for
other longitudinal surveys (EDER and ERMEU).

[Asrul Akmal Shafie]

Thank you all for the quick response. The resounding answer seems to be
that SPSS is unable to handle recurrent event counting process. The
problem with me is that this is the only statistical software subscribed
by my department.

Restructuring the data seems a bit more advance to me but I will try to
locate the paper now.

R seems to be an interesting option. But Deborah, what do you mean by R
not able to write to disk? Does it mean not able to save the data or
output into hard disc?

Many thanks for your kind advice.

Asrul A Shafie
Student
Centre for Socioeconomic Research



On Wed, 1 Nov 2006 19:26:37 -0600, Butler, Deborah {FLNA}
<[hidden email]> wrote:

> It also accepts Matlab code.  You could start here in learning it:
>http://www.r-project.org/  Just get in and start using it!
>
>Just a word of caution, R uses memory only and does not write to disk,
>therefore, it is NOT helpful for those of us with large datasets and
>large data manipulations.  This is one of the reasons why a professional
>package such as SPSS is worth the money!  I like the versatility of it,
>however, and plan to use it concurrently with SPSS for special projects.
>I also like that it is free. ;)
>
>
>Debbie Butler
>SCP Modeling Analyst
>Frito Lay Operations
>
>-----Original Message-----
>From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
>Lisa Stickney
>Sent: Wednesday, November 01, 2006 2:58 PM
>To: [hidden email]
>Subject: Re: Recurrent Event Counting Process
>
>Hi Edward,
>
>   A few years ago I was using both S+ & R (the open source version of
>S+),
>and I found "The Basics of S-Plus" by Krause & Olson to be useful for a
>beginner. HTH.
>
>    Best,
>        Lisa
>
>Lisa T. Stickney
>Ph.D. Candidate
>The Fox School of Business
>     and Management
>Temple University
>[hidden email]
>
>
>
>----- Original Message -----
>From: "Edward Boadi" <[hidden email]>
>To: <[hidden email]>
>Sent: Wednesday, November 01, 2006 3:10 PM
>Subject: Re: Recurrent Event Counting Process
>
>
>> What book will you recommend for somebody who want to learn R ?
>>
>> -----Original Message-----
>> From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf
>Of
>> Butler, Deborah {FLNA}
>> Sent: Wednesday, November 01, 2006 1:30 PM
>> To: [hidden email]
>> Subject: Re: Recurrent Event Counting Process
>>
>>
>> If SPSS cannot handle it, I would suggest R.  It's a freeware package
>-
>> very powerful.  You'll need to download a specific package, but I've
>no
>> doubt it could handle it.
>>
>> http://www.r-project.org/
>>
>>
>> Debbie Butler
>> SCP Modeling Analyst
>> Frito Lay Operations
>> 972.334.3934
>>
>> -----Original Message-----
>> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf
>Of
>> Gene Maguin
>> Sent: Wednesday, November 01, 2006 9:43 AM
>> To: [hidden email]
>> Subject: Re: Recurrent Event Counting Process
>>
>> Asrul,
>>
>> The simple answer No. You can't run this kind of analysis directly.
>That
>> said, I wonder if spss can't be made to run it. My idea is based on
>the
>> fact
>> that a discrete time survival analysis can be restructured to be
>solved
>> using ordinary logistic regression. At least one reference for this is
>> Paul
>> Allison. It seems to me that a competing risks model might be
>> restructured
>> to be solvable using either ordinal logistic regression or multinomial
>> logistic regression. However, I have never seen anybody do this nor
>have
>> I
>> read about doing this. I'm not a statistician so there's lots of
>things
>> I
>> don't know about this. If you haven't gotten a reply perhaps somebody
>> who
>> has done this kind of model would reply to the list. I think that
>Hector
>> had
>> a problem that involved this kind of analysis a while back.
>>
>> All that said, you can do this sort of problem using other software,
>> notably, Mplus, Mlwin, perhaps, HLM or SAS. You might want to look at
>> Judith
>> Singer's book, Applied longitudinal data analysis to see if she has
>> anything
>> about this sort of analysis in there. There may also be free software
>> that
>> can do this analysis. I'd investigate Hedeker's programs.
>>
>> Good luck, Gene Maguin
>>