NL Combining RP & SP data
44 views
Skip to first unread message
R Morales
Jan 25, 2025, 4:47:02 AM1/25/25
to Biogeme
Let’s say I am trying to estimate a model with three alternatives: bus, train, and car, with one nest for the car and another for public transport. How should I handle this when I have SP and RP data? Should I create four nests (PT - SP, PT - RP, Car - SP, Car - RP)? Or should I first create a level with two nests for PT and Car, and then a second level for RP/SP? Could you please direct me to an example model file or references for combining RP-SP data sources to estimate NL models? Thanks in advance
R Morales
Jan 28, 2025, 10:31:44 AM1/28/25
to Biogeme
I’d appreciate it if anyone could help me with this, please.
Michel Bierlaire
Jan 28, 2025, 11:32:57 AM1/28/25
to rocio....@diadro.com, Michel Bierlaire, Biogeme
The best way to think about it is to start by developing first a model for RP data, and, independently, a separate model for SP data. It can be anything (logit, nested, cross-nested, etc.)
Then, when you pull the data together, you declare the SP alternatives unavailable for the RP observations, and the other way around.
> On 17 Jan 2025, at 16:17, R Morales <rocio....@diadro.com> wrote:
>
> Let’s say I am trying to estimate a model with three alternatives: bus, train, and car, with one nest for the car and another for public transport. How should I handle this when I have SP and RP data? Should I create four nests (PT - SP, PT - RP, Car - SP, Car - RP)? Or should I first create a level with two nests for PT and Car, and then a second level for RP/SP? Could you please direct me to an example model file or references for combining RP-SP data sources to estimate NL models? Thanks in advance
>
> --
> You received this message because you are subscribed to the Google Groups "Biogeme" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to biogeme+u...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/biogeme/193e941b-4b78-493e-8754-23b6654c4e85n%40googlegroups.com.
Michel Bierlaire
Transport and Mobility Laboratory
School of Architecture, Civil and Environmental Engineering
EPFL - Ecole Polytechnique Fédérale de Lausanne
http://transp-or.epfl.ch
http://people.epfl.ch/michel.bierlaire
Then, when you pull the data together, you declare the SP alternatives unavailable for the RP observations, and the other way around.
> On 17 Jan 2025, at 16:17, R Morales <rocio....@diadro.com> wrote:
>
> Let’s say I am trying to estimate a model with three alternatives: bus, train, and car, with one nest for the car and another for public transport. How should I handle this when I have SP and RP data? Should I create four nests (PT - SP, PT - RP, Car - SP, Car - RP)? Or should I first create a level with two nests for PT and Car, and then a second level for RP/SP? Could you please direct me to an example model file or references for combining RP-SP data sources to estimate NL models? Thanks in advance
>
> You received this message because you are subscribed to the Google Groups "Biogeme" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to biogeme+u...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/biogeme/193e941b-4b78-493e-8754-23b6654c4e85n%40googlegroups.com.
Michel Bierlaire
Transport and Mobility Laboratory
School of Architecture, Civil and Environmental Engineering
EPFL - Ecole Polytechnique Fédérale de Lausanne
http://transp-or.epfl.ch
http://people.epfl.ch/michel.bierlaire
Reply all
Reply to author
Forward
0 new messages