Jump to content

Talk:Endogeneity (econometrics)

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


First sentence incorrect

[edit]

A variable co-varying (correlation implies linearity) with variance in the error term describes heteroskedasticity, NOT endogeneity. — Preceding unsigned comment added by 207.38.229.133 (talk) 02:14, 25 September 2014 (UTC)[reply]

Merge?

[edit]

Any reason not to merge this with endogenous? Pdbailey (talk) 02:00, 8 May 2008 (UTC)[reply]

Merging is probably a good idea. In the same action, I recommend splitting "endogenous (economics)" out of the endogenous article. My only hesitancy is that there's not much information on them right now, though there should be. -FrankTobia (talk) 16:19, 16 September 2008 (UTC)[reply]
I did a main article link from endogenous (that is a disambig of sorts) and merged the econometrics article in. Pdbailey (talk) 17:59, 16 September 2008 (UTC)[reply]

Example

[edit]

The example given in the lead doesn't seem to fit the definition. The example is endogenous "because producers change their price in response to demand and consumers change their demand in response to price". How is this related to "a correlation between the parameter or variable and the error term"? Yaris678 (talk) 14:50, 13 December 2009 (UTC)[reply]


The focus

[edit]

The focus should be on exogeneity rather than endogeneity. At present, it is weird that there is no article named exogeneity, but we have endogeneity. But they can be one article, and the title should be the exogeneity analysis.

Then we define the usual exogeneity, sequential exogeneity, strong\strict exogeneity, if a variable is not exogenous for any parameter, then it is endogenous. Jackzhp (talk) 13:25, 24 February 2011 (UTC)[reply]

Economics

[edit]

I don't see what this has to do with economics as opposed to applied statistics. Correlation with the error term is not a concept/problem unique to models utilizing economic data, so this seems to be an example of confusion between econometrics and statistics more generally. I'd like to suggest eliminating all references to economics and econometrics that do not specifically refer to economic data/issues. Gwhitfield(talk) 17:00, 8 June 2012 (UTC)[reply]

That is true. “Endogenous” in any economic subfield other than econometrics means “a variable that is determined from within a simultaneous equations model”, rather than being given (“exogenous”). For instance, see endogenous money, or endogenous growth theory. Endogeneity, as explained in this article, refers to the existence of an endogenous explanatory variable in an econometric model. I'll move the article. --bender235 (talk) 18:37, 25 November 2013 (UTC)[reply]

Move discussion in progress

[edit]

There is a move discussion in progress on Talk:Exogenous and endogenous variables which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 17:34, 2 August 2020 (UTC)[reply]