Timeline for What do "endogeneity" and "exogeneity" mean substantively?
Current License: CC BY-SA 3.0
18 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 23, 2022 at 20:07 | answer | added | Aksakal | timeline score: 1 | |
| Nov 12, 2017 at 15:00 | history | protected | kjetil b halvorsen♦ | ||
| Nov 12, 2017 at 14:57 | history | edited | kjetil b halvorsen♦ |
edited tags
|
|
| Sep 1, 2016 at 3:26 | comment | added | John | This does not provide an answer to the question. To critique or request clarification from an author, leave a comment below their post. - From Review | |
| Aug 31, 2016 at 21:49 | comment | added | Andy | The answer is also not technically correct regarding the terminology. The estimator (I assume Ivan talks about OLS here) and test statistics are still "usable" in the sense that they are working, i.e. they can be applied. What you don't get is an unbiased estimate(!) of a certain population parameter - if you were looking for any such parameter to begin with, as pointed out by Matthew. | |
| Aug 31, 2016 at 20:37 | comment | added | whuber♦ | @Matthew: fair enough; that is helpful and constructive criticism. | |
| Aug 31, 2016 at 20:33 | comment | added | Matthew Drury | @whuber I don't know, it's so short I can't really tell. But I was thinking, for example, that the estimated model can be useful for prediction (or just association) even if you have endogeneity, so "no longer has usable estimators" seems false without clarification. | |
| Aug 31, 2016 at 20:30 | comment | added | whuber♦ | @Matthew It seems to me this post does attempt to respond to the question "what does this mean in a real world sense?" It would be nice to see the explanation fleshed out so that people could appreciate it better. | |
| Aug 31, 2016 at 19:41 | comment | added | Matthew Drury | I agree with @gung, and would like to emphasise that a complete answer would address "Usable for what purpose"? Many of the above answers deal with this question very well. | |
| Aug 31, 2016 at 19:07 | comment | added | Ivan | When you have endogeneity your regression no longer has usable estimators or test statistics. | |
| Apr 21, 2014 at 19:17 | answer | added | bearvarine | timeline score: 11 | |
| Sep 14, 2013 at 23:31 | comment | added | gung - Reinstate Monica |
All three of the answers below are very good (+1 to each). If you want another source of information, I discuss this topic here: Estimating $b_1x_1+b_2x_2$ instead of $b_1x_1+b_2x_2+b_3x_3$, & illustrate it w/ a simulation in R.
|
|
| Sep 14, 2013 at 23:11 | history | edited | gung - Reinstate Monica | CC BY-SA 3.0 |
shortened title; removed peripheral comments; edited
|
| May 21, 2013 at 15:38 | answer | added | Bill | timeline score: 85 | |
| May 21, 2013 at 10:27 | answer | added | generic_user | timeline score: 9 | |
| May 21, 2013 at 7:35 | answer | added | JohnRos | timeline score: 29 | |
| May 21, 2013 at 6:40 | review | First posts | |||
| May 21, 2013 at 7:39 | |||||
| May 21, 2013 at 6:22 | history | asked | user25901 | CC BY-SA 3.0 |