Statsmodels: Discrete choice models
by Ana Martinez Pardo for Python Software Foundation
The aim of this project is to add discrete choice models to statsmodels and fill a gap in the set of discrete models that are currently available. We want to bring to the comunity a BSD licensed tool, which based on hypothetical or real world choices, can study the causes of the choice among alternatives and forecast market behavior. This project is intended, first, work on the currently implemented Multinomial Logit and the Nested Logit algorithms and, then, implement Mixed Logit algorithms. Also,it is proposed to implement flexible model specification and several supporting functions for the summary of the model, the statistics result, and statistical tests to check heteroscedasticity, the nesting structures and random parameters of the model. Implement Mixed Logit model will do possible study complex markets and, provide a user friendly way, define complex discrete choice models. This will catapult to statsmodels to rank between the best tools for discrete choice model estimation.