Semi-Parametric Estimation of Microeconometric Models with Qualitative and Limited Dependent Variables

鉴于现有的估计技术，估计生产函数和其他微观模型充满了统计困难，即使在线性模型的情况下也是如此。对于非线性模型，现有技术的可靠性甚至更低。本项目将分析一些微观计量经济学模型，包括单方程模型、方程组和联立方程模型，所有这些模型都具有定性或有限因变量。主要目标是建立清晰的识别条件，建立一致性，以及估计量的渐近正态性。此外，估计器的有限样本性能将通过蒙特卡罗研究和将其应用于实际数据进行调查。这个项目的目的是发展一种半参数方法估计微观计量模型。这项研究很重要，因为半参数方法的发展提供了一种更好地估计大类非线性模型的方法。这种估计方法应该找到各种应用的经济学家感兴趣的实证研究。

Estimating production functions and other micro models is fraught with statistical difficulties given existing estimating techniques, even in the case of linear models. With nonlinear models, the reliability of existing techniques is even less. This project will analyze a number of microeconometric models, including single equation models, systems of equations and simultaneous equation models, all with qualitative or limited dependent variables. The major goals are to establish clear identification conditions, to establish consistency, and asymptotic normality of the estimators. In addition, the finite sample performance of the estimators will be investigated both by Monte Carlo studies and by applying them to actual data. The purpose of this project is to develop a semi-parametric method for the estimation of microeconometric models. This study is important because the development of semi-parametric methods offers a way to better estimate a large class of non-liner models. This estimation method should find a variety of applications by economists interested in empirical research.