Estimation of Discrete Choice Models Using DCM for Ox

Abstract

DCM (Discrete Choice Models) is a package for estimating a class of discrete choice models. Written in Ox, DCM is a class that implements a wide range of discrete choice models including standard binary response models, with notable extensions including conditional mixed logit, mixed probit, multinomial probit, and random coefficient ordered choice models. The current version can handle both cross-section and static panel data. DCM represents an important development for the discrete choice computing environment in making available a broad range of models which are now widely used by academics and practitioners. Developed as a derived class of Modelbase, users may access the functions within DCM by either writing Ox programs which create and use an object of the DCM class, or use the program in an interactive fashion via OxPack in GiveWin. We demonstrate the capabilities of DCM by using a number of applications from the discrete choice literature.