Limited Dependent Variable Models with Fixed Effects

具有固定效应的有限因变量模型

• 批准号: 9009879
• 负责人: Bo Honore
• 金额: $ 6.76万
• 依托单位: Northwestern University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1990
• 资助国家: 美国
• 起止时间: 1990-07-01 至 1992-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9009879&HistoricalAwards=false
• 关键词: Limited; Dependent; Variable; Models; Fixed

The project will develop estimators of limited dependent variable models, also known as Tobit models, with fixed effects. Individually, fixed effects models and Tobit models are frequently used in empirical economics. A Tobit model with fixed- effects is implied by a structural life cycle model of labor supply and many other models, and is of interest as a reduced form specification. It is desirable to have asymptotic results for fixed effects estimators as the number of individuals increases (with the number of observations per individual fixed). No current estimators for Tobit models with fixed effects are asymptotically normal, or even consistent, as the number of individuals gets large. The project will solve this problem and at the same time relax some of the distributional assumptions. The efficiency of the estimators that are constructed will also be examined. The research will be supplemented with simulations. Given the availability of important panel data sets in economics, the development of these estimators will improve the estimation of many important types of economic models, especially in the area of labor economics.

该项目将开发具有固定效应的有限因变量模型（也称为Tobit模型）的估计器。固定效应模型和Tobit模型分别是实证经济学中常用的两种模型。具有固定效应的Tobit模型是由劳动力供给的结构生命周期模型和许多其他模型所隐含的，并且是一种简化形式规范。随着个体数量的增加（每个个体的观测数量固定），固定效应估计量的渐近结果是可取的。当个体数量变大时，具有固定效应的Tobit模型的当前估计量没有一个是渐近正态的，或者甚至是一致的。该项目将解决这个问题，同时放宽一些分配假设。所构造的估计器的效率也将被检查。这项研究将辅以模拟。考虑到经济学中重要的面板数据集的可用性，这些估计器的开发将改善许多重要类型的经济模型的估计，特别是在劳动经济学领域。