Timeline for Asymptotic normality of random vector
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 7, 2019 at 12:00 | history | tweeted | twitter.com/StackStats/status/1170305848113029121 | ||
| Aug 15, 2019 at 3:48 | vote | accept | user143487 | ||
| Aug 14, 2019 at 12:05 | history | edited | COOLSerdash | CC BY-SA 4.0 |
Some very minor typographical improvements and a tag added.
|
| Aug 13, 2019 at 14:08 | history | edited | whuber♦ |
edited tags
|
|
| Aug 13, 2019 at 14:08 | answer | added | whuber♦ | timeline score: 2 | |
| Aug 12, 2019 at 17:08 | answer | added | Alecos Papadopoulos | timeline score: 1 | |
| Jun 30, 2019 at 6:46 | comment | added | user143487 | I have a parameter vector for which I have a vector of unbiased estimators. Each estimator in the latter vector can be expressed as a linear combination of the means of i.i.d. r.v's with finite variances. So by CLT, each estimator is asymptotically normal as the number of observations becomes large. However, these estimators are correlated. I want to know if anything can be said about the asymptotic distribution of the random vector of estimators. Do I have to consider multivariate CLT for a sequence of such vectors ? | |
| Jun 30, 2019 at 3:07 | comment | added | Glen_b | Can you clarify the situation (how the CLT applies)? It's certainly possible to have situations where components are exactly normal and the distribution is not jointly normal, so it's important to be clearer about what the situation is here. | |
| Jun 29, 2019 at 5:15 | comment | added | BruceET | To the extent that the elements of your $n$-vector are normal with mean $\mu$, the vector should be approximately $n$-variate normal with means $\mu$ and the known variance-covariance matrix. | |
| Jun 29, 2019 at 4:27 | history | asked | user143487 | CC BY-SA 4.0 |