[R] clogit, modeling change...

Stefan Petersson stefan at inizio.se
Fri May 25 09:30:59 CEST 2012


Hi,

I've set up a discrete choice experiment on magazine preference
following the guidelines from Azaiki. H & Nishimura K. "Design and
Analysis of Choice Experiments Using R. A Brief Introuction" from
Agricultural Informatin Research 17(2), 2008. 86-94.

I'm working with the clogit() function from the survival package. I've
set up the data like this:

Person Id / Choice Set / Stated Preference ... (the dots being
magazine characteristics with price, print quality etc)
1 1 0 ...
1 2 0 ...
1 3 1 ...
2 1 0 ...
2 2 1 ...
2 3 0 ...
3 1 1 ...
3 2 0 ...
3 3 0 ...
...   ...

The clogit() gives me coefficients etc. However, the literature states
that one can model change in, say, price. Answering questions like
"How many will read magazine 2 when price falls with $2"?. Can someone
please outline a simple example on how to calculate / model such a
change?

TIA

 // stefan



More information about the R-help mailing list