Hurdle and non-hurdle specifications of finite mixture count data models for medical care demand

DOI IR Web Site

Search this article

Abstract

This paper considers finite mixture count data models for medical care demand using hurdle and non-hurdle specifications. The National Medical Expenditure Survey (NMES) example shows that the standard finite mixture negative binomial model is not always the best choice. Moreover, we conclude that the performance of the models considered in this paper is relatively good.

Journal

Related Projects

See more

Keywords

Details 詳細情報について

Report a problem

Back to top