作者: Erik Biørn , Jayalakshmi Krishnakumar
DOI: 10.1007/978-3-540-75892-1_10
关键词: Observational error 、 Panel data 、 Econometrics 、 Endogeneity 、 Residual 、 Simultaneity 、 Regression analysis 、 Term (time) 、 Object (philosophy) 、 Mathematics
摘要: This chapter is concerned with the problem of endogeneity certain explanatory variables in a regression equation. There are two potential sources panel data model individual and time specific effects : (i) correlation between (when treated random) (ii) residual/idiosyncratic error term. The first case was extensively dealt Chapter 4 this book hence we will not go into it here. In more non-zero overall consisting both effect genuine disturbance One might call double as opposed to single former situation. consider major causes encountered practical situations. them presence measurement errors variables. be object study Section 9.2. Another source simultaneity that arises when equation one several structural equations simultaneous contains current endogenous 9.3 look detail 9.4 concludes