Properties of Estimators under Standard Assumptions
In this section we shall discuss the properties of various estimators of the T obit model under the assumptions of the model. The estimators we shall consider are the probit maximum likelihood (ML), least squares (LS), Heckman’s two-step least squares, nonlinear least squares (NLLS), nonlinear weighted least squares (NLWLS), and Tobit ML estimators.
Income, marital status, number of children Years of schooling, working experience
Sex, age, number of years married, number of children, education, occupation, degree of religiousness Preprogram hours worked, change in the wage rate, family characteristics Ratio of social security benefits lost at time of full-time employment to full-time earnings Price of contributions, income
Wages of husbands and wives, education of husbands and wives, income
Earnings before the program, husband’s and wife’s education, other family characteristics, unemployment rate, seasonal dummies Research expenditure of the pharmaceutical industry, stringency of government regulatory standards
Accumulated work release funds, number of months after release until first job, wage rate after release, age, race, drug use