Timeline for Dimension reduction
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 28, 2019 at 17:15 | comment | added | user0735 | When k>1 I still only require r restrictions for identification rather than rk, I am thinking whether a SVD on M would somehow get me the final sums in F in terms of r and N that I need, rather than rk and Nk. | |
| Apr 28, 2019 at 17:12 | history | edited | user0735 | CC BY-SA 4.0 |
added 2 characters in body
|
| Apr 28, 2019 at 17:04 | history | edited | user0735 | CC BY-SA 4.0 |
added 20 characters in body
|
| Apr 28, 2019 at 17:02 | comment | added | user0735 | Thanks Federico. The formula for F is true and I can confirm it holds when k=1. The restrictions are required for identification purposes. The eigenvalues are those of the G matrix. | |
| Apr 28, 2019 at 16:46 | comment | added | Federico Poloni | This question is difficult to parse. What do you mean by "end up with"? What reasons do you have to suspect that that formula for $F$ is true? What is a "restriction"? Who imposes it? The $\lambda_n$ presumably are eigenvalues of some matrix, but which one? | |
| Apr 28, 2019 at 16:36 | history | asked | user0735 | CC BY-SA 4.0 |