|This is the talk page for discussing improvements to the Omitted-variable bias article.
This is not a forum for general discussion of the article's subject.
|WikiProject Statistics||(Rated Start-class, Mid-importance)|
I really feel like, while accurate, the description on this page isn't helpful. Perhaps a plain english explanation, followed by 'more specifically, blah blah'. Editing accordingly Odoketa (talk) 03:53, 21 February 2011 (UTC)
About the "This article may be confusing or unclear to readers" warningEdit
I was going through the Key Terms and Concepts in Chapter 17 of Econometric Analysis (Greene, 2011) by looking up each term in Wikipedia, and this is the most helpful article among the bunch. Maybe we should get rid of the warning. dbabbitt (talk) 12:51, 6 October 2013 (UTC)
Dr. Ashenfelter's comment on this articleEdit
Dr. Ashenfelter has reviewed this Wikipedia page, and provided us with the following comments to improve its quality:
I would say this is a pretty good entry, material for the non-technical reader and the more technical reader. The references to Wooldrige and Greene are standard, but I'm not familiar with the other two.
We hope Wikipedians on this talk page can take advantage of these comments and improve the quality of the article accordingly.
Dr. Ashenfelter has published scholarly research which seems to be relevant to this Wikipedia article:
- Reference : Ashenfelter, Orley & Greenstone, Michael, 2004. "Estimating the Value of a Statistical Life: The Importance of Omitted Variables and Publication Bias," IZA Discussion Papers 1037, Institute for the Study of Labor (IZA).
I think that this article is not very accessible and is also highly focused on the econometric side of things.
Gauss Markov Section Needs WorkEdit
The section on the Gauss Markov isn't quite right, or at least is misleading. Suppose we have Y=a+XB+e as our short regression and Y=a+XD+ZG+n. With an omitted variable, the estimate of the short regression remains consistent. That is, the short regression still reflects the correct expectation E[Y|X] and E[e|X]=0. Its just that the short coefficient B is an inconsistent estimator of the coefficient D in the long regression. Clinton.tepper (talk) 02:28, 15 October 2018 (UTC)