Market shares and ASC (RP & SP)
R Morales
I am trying to calibrate a logit model using RP & SP data. Transport alternatives are the same in both data sets. The state of practice recommends always use RP-ASC for existing alternatives and since only RP data represent observations of behaviour that have actually taken place, then estimate a joint RP/SP model using specific RP/SP ASC to determine values for all the model coefficients, then adjusting the RP-ASC for the existing alternatives using the RP data and constrain the other coefficients to values estimated in the model calibration process to match the base year observations.
However, what happens if the market shares of the RP sample differ from those observed in the real market (as determined from real data such as public transport operator statistics, traffic counts, OD matrices, etc.)? How do I take this into account? Should the RP sample be expanded to reproduce the real market shares? What procedure should be followed in such cases?
Thanks in advance
Michel Bierlaire
https://eml.berkeley.edu/books/choice2.html
> 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/ad792e1d-1195-4824-a762-1bff7171c511n%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