The integer programing extreme value (IPEV) model accounts the integer property of trip data. The IPEV model is consistent with utility theory and provides a single structural framework for simultaneously modeling the choice of alternatives and quantity decisions with the constraint of the integer value of consumption. This model has a closed-form probability expression. This sample files include the estimation code using closed-form or simulation, the demand prediction, the welfare analysis, and the sample dataset.
Figure 1. Continuous and integer demand models.
Estimation:
- Set the parameters in IPEV.gss (required maxlik) or IPEVmt.gss (required maxlikmt)
- U_FUNCTION: the utility function
- For the estimation with closed-form, set SIMULATE = 0
- For the estimation with simulation, set SIMULATE = 1
- Run the estimation code in the GAUSS:
>> run IPEV.gss
or
>> run IPEVmt.gss
Prediction:
- Set the parameters in DEMAND.gss
- U_FUNCTION: the utility function
- B: vector of estimated parameters
- VCOV: variance-covariance matrix
- Set secnarios for prediction and welfare analysis.
- Run the prediction code in the GAUSS:
>> run DEMAND.gss
For the detail, see manual.pdf.
MIT License