Timeline for Why is the rank of covariance matrix at most $n-1$?
Current License: CC BY-SA 4.0
2 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 31, 2022 at 21:26 | comment | added | whuber♦ | But you don't need two! It depends on what you assume about that variable; and even then, you can always estimate the variance based on one value. There's going to be an issue of how good the estimate might be, but that's not really germane. | |
| Mar 31, 2022 at 21:18 | history | answered | Robert Bray | CC BY-SA 4.0 |