工具变量两阶段最小二乘法.pptVIP

Instrumental variables 2SLS 工具变量与两阶段最小二乘法 ◆y=+Bx1+月2x2+..员xk+ 兀0+兀+ Intermediate Econometrics Yan Shen Chapter Outline 本章提要 Omitted Variables in a simple regression model 简单回归中的遗漏变量 IV estimation of the Multiple Regression 多方程回归中的工具变量估计 a Two Stage Least squares 两阶段最小二乘法 IV solutions to errors-in-variables problem 用工具变量解决测量误差问题 a Testing for endogeneity 检验内生性 Intermediate Econometrics Yan Shen 2 ecture Outline 本课提要 a Motivation: Why using Iv? 出发点:为何用工具变量 Statistical Inference with the iv estimator IV估计中的统计推断 ■ Properties of iv with a poorⅣV “坏”工具变量的性质 a Computing r squares after Iv 计算ⅠV估计的R方 a IV estimation of the multiple regression model 多方程回归的ⅣV估计 Intermediate Econometrics Yan Shen Problem to start with 从这个问题出发 a If important variables are omitted what should we do? 如果一些重要的变量被遗漏,我们应当怎 么办? Intermediate Econometrics Yan Shen The ways out 些办法 gnore the proble, pretend that it does not exist 忽略这个问题,假装这个问题并不存在 a Find and use a suitable proxy 使用代理变量 a Uses an estimation method that recognizes the presence of the omitted variable 使用一种对遗漏变量稳健的估计方法。 Intermediate Econometrics Yan Shen Why Use Instrumental variables? 为何使用工具变量? Instrumental Variables (Iv estimation is used when your model has endogenous X s 当模型解释变量具有内生性时,使用工具 变量估计 ■ That is, when cov(x,u)≠O 即,Co(x,u)≠O时 Intermediate Econometrics Yan Shen Why Use Instrumental variables? 为何使用工具变量? Thus. iv can be used to address the problem of omitted variable bias 所以,Ⅳ可以用来解决遗漏变量偏差 a additionally, iv can be used to solve the classic errors-in-variables problem 而且,Ⅳ可用来解决经典的测量误差问题 Intermediate Econometrics Yan Shen 7 Instrumental Variable: Who qualifies? 什么样的变量可以作为IV? In order for a variable, z, to serve as a valid instrument for x, the following must be true 针对内生变量x的一个有效的工具变量z应当满 足如下条件 The instrument must be exogenous 工具变量应为外生 a That is, Cov(Z, u=0(15.4) 即CoV(z,l)=O Intermediate Econometrics Yan Shen Instrumental Variable: Who qualifies? 什么样的变量可以作为ⅣV? The instrument must be correlated with the endogenous variable x 工具变