Elements of Modeling with Mixed Variables

Farrokh Guiahi 
(Department of IT/QM, Zarb School of Business, Hofstra University, Hempstead NY 11549, USA)

Abstract: This paper discusses briefly models for mixed variables. In particular, suitable joint distribution of mixed variables is provided by reference to specific examples. Procedures based on model comparison are utilized to study the dependency structure pertaining to a categorical and a continuous variable. Estimation of parameters and computation of Likelihood function is addressed by providing the necessary code as an ADMB program.

Key words: models for mixed variables; ADMB program

JEL code: C18