Browsing by Subject "211"

Now showing 1 - 7 of 7
  • Thumbnail Image
    Item
    Amemiya's Generalized Least Squares and Tests of Overidentification in Simultaneous Equation Models with Qualitative or Limited Dependent Variables
    (Center for Economic Research, Department of Economics, University of Minnesota, 1991-05) Lee, Lung-Fei
    Amemiya's generalized least squares method for the estimation of simultaneous equation models with qualitative or limited dependent variables is known to be efficient relative to many popular two-stage estimators. This note points out that test statistics for overidentification restrictions can be obtained as by-products of Amemiya's generalized least squares procedure. Amemiya's procedure is shown to be a minimum chi-square method. The Amemiya procedure is valuable both for efficient estimation and for model evaluation of such models.
  • Thumbnail Image
    Item
    Efficient Semiparametric Scoring Estimation of Sample Selection Models
    (Center for Economic Research, Department of Economics, University of Minnesota, 1990-02) Lee, Lung-Fei
    A semi parametric profile likelihood method is proposed for estimation of sample selection models. The method is a two step scoring semi parametric estimation procedure based on index formulation and kernel density estimation. Under some regularity conditions, the estimator is asymptotically normal. This method can be applied to estimation of general sample selection models with multiple regimes and sequential choice models with selectivity. For the binary choice sample selection model, the estimator is asymptotically efficiency in the sense that its asymptotic variance matrix attains the asymptotic bound of G. Chamberlain.
  • Thumbnail Image
    Item
    On Efficiency of Methods of Simulated Moments and Maximum Simulated Likelihood Estimation of Discrete Response Models
    (Center for Economic Research, Department of Economics, University of Minnesota, 1990-09) Lee, Lung-Fei
    This article has considered methods of simulated moments for estimation of discrete response models. We have introduced a modified method of simulated moments of McFadden [1989]. Using the same number of Monte Carlo draws as in McFadden's method of simulated moments, our estimator is asymptotically efficient relative to McFadden's estimator. In addition to the method of simulated moments, we have considered also maximum simulated likelihood estimation methods. The estimators are shown to be consistent and asymptotically normal without excessive number of Monte Carlo draws.
  • Thumbnail Image
    Item
    Semiparametric Instrumental Variable Estimation of Simultaneous Equation Sample Selection Models
    (Center for Economic Research, Department of Economics, University of Minnesota, 1991-06) Lee, Lung-Fei
    The identification and estimation of a semiparametric simultaneous equation model with selectivity have been considered. The identification of structural parameters from reduced form parameters in the semi parametric model requires stronger conditions than the usual rank condition in the classical simultaneous equation model or the parametric simultaneous equation sample selection model. The necessary order condition for identification in the semiparametric model corresponds to the over-identification condition in the classical model. Semiparametric two-stage estimation methods which generalize the two-stage least squares method and the generalized two-stage least squares method for the parametric model are introduced. The semi parametric generalized least squares estimator is shown to be asymptotically efficient in a class of semiparametric instrumental variable estimators.
  • Thumbnail Image
    Item
    Semiparametric Maximum Profile Likelihood Estimation of Polytomous and Sequential Choice Models
    (Center for Economic Research, Department of Economics, University of Minnesota, 1989-12) Lee, Lung-Fei
    This article considers semiparametric estimation of discrete choice models. The estimation methods are some semiparametric maximum profile likelihood methods which generalize Klein and Spady [1987] to the estimation of polytomous choice and sequential choice models. Special emphases are on the correction of asymptotic bias and negative density estimates caused by high order kernel density estimation. The estimators are shown to be Vn consistent and asymptotically normal. They attain the asymptotic efficiency bound of semiparametric estimation with some infinite dimensional parameter spaces of index probability functions.
  • Thumbnail Image
    Item
    Semiparametric Nonlinear Least Square Estimation of Truncated Regression Models
    (Center for Economic Research, Department of Economics, University of Minnesota, 1990-02) Lee, Lung-Fei
    This article provides a semi parametric method for the estimation of truncated regression models where the disturbances are independent with the regressors before truncation. This independency property provides useful information on the identification and estimation of the model. Our estimate is shown to be Vn-consistent and asymptotically normal. Consistent estimate of the asymptotic covariance matrix of the estimator is provided. Monte Carlo experiments are performed to investigate some finite sample properties of the estimator.
  • Thumbnail Image
    Item
    Semiparametric Two Stage Estimation of Sample Selection Models Subject to Tobit-Type Selection Rules
    (Center for Economic Research, Department of Economics, University of Minnesota, 1990-07) Lee, Lung-Fei
    A semiparametric two stage estimation method is proposed for the estimation of sample selection models which are subject to Tobit-type selection rules. With randomization restrictions on the disturbances of the model, all the regression coefficients in the model are in general identifiable without exclusion restrictions. The proposed estimator is shown to be Vn-consistent and asymptotically normal. Some Monte Carlo results to demonstrate its finite sample performance are also provided.