Limited Dependent Variable Models
The standard regression model assumes that the dependent variable, Y, is measured quantitatively. The independent (or regressor) variables, Xi, may be measured quantitatively or qualitatively. A dummy regressor is a variable that is measured qualitatively. Logit models apply to situations where the dependent variable is dichotomous in nature, taking a 0 or 1 value. For example, the dependent variable, Y, could be whether or not a person is unemployed (“employed” = 1, “unemployed” = 0). The regressors could include X1 the average national wage rate, X2 the individual’s education, X3 the national unemployment rate, X4 family income etc. The question arises as to how we handle models involving dichotomous dependent variables.