ABSTRACT
This chapter presents probit and logit regression models which are used for examining binary decisions and differences between these binary-choice models and the linear regression model. It describes Tobit model which is used for censored dependent variables and self-selection bias when a self-selected choice variable is included as a regressor. The linear probability model is introduced to outline the issues associated with qualitative dependent variables, although it has some theoretical problems and is rarely used. In most applications, the logit and probit models produce similar results. The choice between the two models is one of convenience. The chapter examines whether institutional investors create or reduce incentives for corporate managers to reduce investment in research and development to meet short-term earnings goals. The choice-based models consist of two equations. One equation is for the choice decision, and the other is for the response variable of the main interest.
