Bayesian Conjoint Analysis

[Michael Pearmain-2]

Hi All,
Not a question about SPSS per say, but due to the nature of the list, and
some of the contributors i was looking for some advice and possibly an off
list discussion.

I want to implement a Bayesian Conjoint design, and was looking to see if
any others had approached this area before, or had some sample code, the
best prospect looks like an R module called bayesm.  If other packages are
know i would try those, (though open source is best as we don't have
sawtooth)

Can anyone offer any advice on difficulties or design architecture for this?

Many thanks in advance

Mike

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Anthony Babinec]

The Bayesian part might reside more in the
analysis than in the design. Find a copy
of "Bayesian Statistics and Marketing" by
Peter Rossi, Greg Allenby, and Rob McCulloch.
They are the authors of the BAYESM package.
They also have some published articles.

Anthony Babinec
[hidden email]


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Michael Pearmain
Sent: Friday, December 12, 2008 6:37 AM
To: [hidden email]
Subject: Bayesian Conjoint Analysis

Hi All,
Not a question about SPSS per say, but due to the nature of the list, and
some of the contributors i was looking for some advice and possibly an off
list discussion.

I want to implement a Bayesian Conjoint design, and was looking to see if
any others had approached this area before, or had some sample code, the
best prospect looks like an R module called bayesm.  If other packages are
know i would try those, (though open source is best as we don't have
sawtooth)

Can anyone offer any advice on difficulties or design architecture for this?

Many thanks in advance

Mike

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Zetu, Dan]

I second Anthony's comment. The design of the conjoint study has nothing
to do with the Hierarchical Bayes method of computing individiual-level
utilities (prior to this capability, preferences and attribute
importances could be computed only in an aggregate fashion). As Anthony
mentioned, this method is used in the analysis phase. Another good
source of information on Hierarchical Bayes is the Sawtooth software web
site - Working papers.

-------------------------------
Dan Zetu
Analytical Consultant
R. L. Polk & Co.
248-728-7278
[hidden email]

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Anthony Babinec
Sent: Friday, December 12, 2008 7:48 AM
To: [hidden email]
Subject: Re: Bayesian Conjoint Analysis

The Bayesian part might reside more in the
analysis than in the design. Find a copy
of "Bayesian Statistics and Marketing" by
Peter Rossi, Greg Allenby, and Rob McCulloch.
They are the authors of the BAYESM package.
They also have some published articles.

Anthony Babinec
[hidden email]


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Michael Pearmain
Sent: Friday, December 12, 2008 6:37 AM
To: [hidden email]
Subject: Bayesian Conjoint Analysis

Hi All,
Not a question about SPSS per say, but due to the nature of the list,
and
some of the contributors i was looking for some advice and possibly an
off
list discussion.

I want to implement a Bayesian Conjoint design, and was looking to see
if
any others had approached this area before, or had some sample code, the
best prospect looks like an R module called bayesm.  If other packages
are
know i would try those, (though open source is best as we don't have
sawtooth)

Can anyone offer any advice on difficulties or design architecture for
this?

Many thanks in advance

Mike

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD