Timeline for Independent variable = Random variable?
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 3, 2023 at 22:03 | comment | added | whuber♦ | @MysteryGuy That's part of it. But there cannot be any strong connection between $X$ and the error, either, as expressed by the first bullet point in the "random design" assumptions in this answer. | |
| Nov 3, 2023 at 21:31 | comment | added | MysteryGuy | @whuber Sorry for being "late" on this answer, but if I understand correctly, the assumptions say that the error should be white noise, right ? | |
| Sep 28, 2020 at 14:18 | comment | added | whuber♦ | Aksakal, because your nonstandard use of the term "serial correlation" is causing confusion, please consider changing the term. Simply striking "serial" ought to fix the problem. If you want to be clearer, you should explicitly quantify the subscripts, as in "No correlation: for all $i\ne j, E[\varepsilon_i\varepsilon_j\mid X]=0.$" (Note, too, the removal of the comma, whose presence is probably a typographical error.) | |
| Sep 25, 2020 at 13:46 | comment | added | ManUtdBloke | I have asked a question related to this answer in this post: stats.stackexchange.com/questions/489132/… | |
| Sep 25, 2020 at 13:30 | comment | added | ManUtdBloke | You have $\varepsilon$ in the the first two assumptions, but in the final assumptions they also feature a subscript. So does the final assumption only hold with respect to the regression model applied to sample data, and not to the regression model applied to the population? | |
| Jul 16, 2020 at 19:16 | comment | added | ColorStatistics | +1 Excellent answer. You might as well add that any econometrics textbook out there is likely assuming the Independent Variable to be a R.V. and any statistics textbook out there is either assuming the Independent Variable to be fixed or listing both possibilities but flushing out the experimental context more. | |
| Jul 10, 2020 at 20:13 | comment | added | 24n8 | For your serial correlation notation, I think you meant to write it as a covariance or correlation instead of expectation? I've never seen expectation written like that before with multiple variables -- is it formal? | |
| Jun 11, 2020 at 14:32 | history | edited | CommunityBot |
Commonmark migration
|
|
| Sep 19, 2017 at 13:59 | history | edited | Aksakal | CC BY-SA 3.0 |
added 180 characters in body
|
| Sep 19, 2017 at 13:51 | history | answered | Aksakal | CC BY-SA 3.0 |