This paper presents a structural estimation of a dynamic model of schooling, work, and occupational choice decisions based on 11 years of data from a sample of young men in the 1979 youth cohort of the National Longitudinal Surveys of Labor Market Experience (NLSY). The model incorporates human capital investment theory and allows for the investigation of whether a theoretically restricted model can quantitatively fit observed data patterns. The results show that an extended human capital investment model can accurately fit data on school attendance, work, occupational choices, and wages, and also produce reasonable forecasts of future work decisions and wage patterns. The structural approach provides rigorous interpretations for the estimated parameters, which has important implications. It allows for the estimation of parameters of interest, such as those of technology, and quantifies the effect of altering specific parameters on decisions. It also enables welfare analysis, allowing for the calculation of the distributional consequences of interventions on lifetime wealth and utility. The paper extends previous work by considering self-selection in the three dimensions of schooling, work, and occupational choice. It combines features of the Heckman and Sedlacek (1985) and Willis (1986) models, each of which is an extension of the basic Roy (1951) framework. The model is extended to one in which decision making is sequential and the environment is uncertain. The estimation of the model involves the repeated numerical solution of a discrete-choice, finite-horizon optimization problem, formulated as a dynamic programming problem. The model is implemented using data from approximately 1,400 white males from the NLSY, tracking their decisions from age 16 to 26. The model accounts for the interrelated nature of schooling, work, and occupational choice, and allows for the estimation of the impact of interventions such as college tuition subsidies on subsequent occupational choice decisions. The paper is organized into sections discussing the basic human capital model, data, estimation results of the basic model, and an extended model. The basic model fails to capture some important quantitative features of the data, so an extended model is presented, which conforms better to the data. The extended model allows for a more accurate estimation of parameters, including the impact of skill heterogeneity, the effect of college tuition subsidies, and the relationship between wages and education and experience. The results show that the extended model provides a better fit to the data and offers more accurate predictions of future work and wage patterns.This paper presents a structural estimation of a dynamic model of schooling, work, and occupational choice decisions based on 11 years of data from a sample of young men in the 1979 youth cohort of the National Longitudinal Surveys of Labor Market Experience (NLSY). The model incorporates human capital investment theory and allows for the investigation of whether a theoretically restricted model can quantitatively fit observed data patterns. The results show that an extended human capital investment model can accurately fit data on school attendance, work, occupational choices, and wages, and also produce reasonable forecasts of future work decisions and wage patterns. The structural approach provides rigorous interpretations for the estimated parameters, which has important implications. It allows for the estimation of parameters of interest, such as those of technology, and quantifies the effect of altering specific parameters on decisions. It also enables welfare analysis, allowing for the calculation of the distributional consequences of interventions on lifetime wealth and utility. The paper extends previous work by considering self-selection in the three dimensions of schooling, work, and occupational choice. It combines features of the Heckman and Sedlacek (1985) and Willis (1986) models, each of which is an extension of the basic Roy (1951) framework. The model is extended to one in which decision making is sequential and the environment is uncertain. The estimation of the model involves the repeated numerical solution of a discrete-choice, finite-horizon optimization problem, formulated as a dynamic programming problem. The model is implemented using data from approximately 1,400 white males from the NLSY, tracking their decisions from age 16 to 26. The model accounts for the interrelated nature of schooling, work, and occupational choice, and allows for the estimation of the impact of interventions such as college tuition subsidies on subsequent occupational choice decisions. The paper is organized into sections discussing the basic human capital model, data, estimation results of the basic model, and an extended model. The basic model fails to capture some important quantitative features of the data, so an extended model is presented, which conforms better to the data. The extended model allows for a more accurate estimation of parameters, including the impact of skill heterogeneity, the effect of college tuition subsidies, and the relationship between wages and education and experience. The results show that the extended model provides a better fit to the data and offers more accurate predictions of future work and wage patterns.