Theory and Applications of Micro-econometrics

本课题的主要目的是重新审视微观计量经济分析中常用的统计方法及其应用。首先，我们考察了用于计量经济分析和金融计量经济分析的主要半参数统计方法。我们研究了经验似然（EL）方法。特别是，我们比较了最大经验似然（MEL）方法和广义矩量法（GMM）。后一种方法在计量经济学分析中已广为人知，并且在过去的二十年里，它一直是计量经济学领域中最流行的半参数方法。我们发现在应用中存在MEL方法优于GMM方法的重要情况，并利用渐近展开方法研究了这些估计方法的小样本性质。其次，我们还研究了有限信息极大似然（LIML）方法的经典方法。 关于我们 并发现当工具数较多时，它在微观计量经济模型中具有渐近最优性。深入研究了ME方法、GMM方法、LIML方法和TSLS（Two-Stage Least Squares）方法在小样本情况下的差异。第三，本项目在MEL方法和LIML方法的研究上取得了许多新的成果。本课题的研究成果已在国内外学术会议上进行了详细的报告，并已（或将）在本报告所列的学术论文中进行了报告。总之，我们已经完成了本课题的最重要目标。正式参与该项目的四名成员撰写了许多论文，并激发了大量相关领域的研究人员和一些具有国际学术视野的统计学家，我们感谢文部科学省和日本科学振兴会对我们的研究项目给予慷慨支持。少

The main purpose of this project was to re-examine the existing statistical methods often used in in micro-econometric analysis and their applications.First we have investigated the major semiparametric statistical methods for analyzing econometric analyses and financial econometric analyses. We have investigated the empirical likelihood (EL) approach. In particular, we have compared the maximum empirical likelihood (MEL) method and the generalized method of moments (GMM). The latter method has been well-known in econometric analyses and it has been the most popular semiparametric method in the past two decades within the field of econometrics. We have found that there are important situations in applications where the MEL method dominates the GMM method and we have investigated the small sample properties of these estimation methods by using the asymptotic expansion method.Second, we have also investigated the classical method of the limited information maximum likelihood (LIML) metho … More d and found that it has an asymptotic optimality in the micro-econometric models when the number of instruments is large. We have investigated the small sample differences among the ME method, the GMM method, the LIML method and the TSLS (Two-Stage Least Squares) method in depth.Third, there have been many new results we have obtained under the research efforts of this project on the MEL method and the LIML method. The details of the results under our research project have been reported in domestic as well as international academic meetings and have been (or will be) reported in academic papers listed in this report.In conclusion, we have accomplished the most important objectives of this project. Four members participated in this project officially have written many papers and also stimulated a large number of researchers in the related fields and some statisticians in the academic international perspectives We thank The Ministry of Education, Science, Sports and Culture and Japan Society for the Promotion of Science for giving the generous support to our research project. Less