Timeline for Solving for auto-regression coefficients when covariance is singular?
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 22, 2022 at 22:40 | comment | added | Yaroslav Bulatov | I think it's an issue of Ledoit-Wolf optimizing for different objective. To get coefficients, I divide entries of precision matrix by entries of the diagonal of the precision matrix. Small bias in the precision matrix estimation produces large error in the coefficients. | |
| Jun 22, 2022 at 22:38 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
added 289 characters in body
|
| Jun 21, 2022 at 22:14 | comment | added | Aksakal | Your comparisons are problematic. The better way to compare would be to start with a known lag structure, then generate the set of rank deficient samples from it. Then study properties of different estimates of lag coefficients. Currently you are looking at a single sample so any conclusions you make are going to be highly uncertain | |
| Jun 20, 2022 at 20:38 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
deleted 11 characters in body
|
| Jun 20, 2022 at 20:32 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
added 248 characters in body
|
| Jun 20, 2022 at 20:26 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
added 248 characters in body
|
| Jun 20, 2022 at 20:05 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
added 443 characters in body
|
| Jun 20, 2022 at 18:51 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
deleted 573 characters in body
|
| Jun 20, 2022 at 18:50 | answer | added | Aksakal | timeline score: 0 | |
| Jun 20, 2022 at 18:47 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
added 5 characters in body
|
| Jun 20, 2022 at 18:38 | history | edited | Yaroslav Bulatov | CC BY-SA 4.0 |
added 5 characters in body
|
| Jun 20, 2022 at 18:29 | history | asked | Yaroslav Bulatov | CC BY-SA 4.0 |