Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/325065 
Authors: 
Year of Publication: 
2025
Series/Report no.: 
IZA Discussion Papers No. 18007
Publisher: 
Institute of Labor Economics (IZA), Bonn
Abstract: 
The most widely used textbooks today conflate three distinct population parameters: the population regression function (PRF), the conditional expectation function (CEF), and the causal effect. They also incorrectly suggest, and sometimes state, that the Conditional Mean Zero assumption implies causal interpretation of regression coefficients. I highlight these issues and show that by incorporating new notation these limitations can easily be overcome.
Subjects: 
conditional mean zero
unbiased estimator
causal effects
JEL: 
A22
C18
Document Type: 
Working Paper

Files in This Item:
File
Size
347.05 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.