Timeline for Why did statisticians define random matrices?
Current License: CC BY-SA 3.0
23 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 5, 2016 at 10:44 | history | tweeted | twitter.com/StackStats/status/794852848525774848 | ||
| Nov 5, 2016 at 0:17 | comment | added | Aksakal | @Hurkyl, by your logic "random forest" could also completely naturally follow from "random" and "forest" | |
| Nov 4, 2016 at 13:13 | vote | accept | Eduardo | ||
| Nov 4, 2016 at 12:51 | comment | added | user41979 | "Random matrix" follows completely naturally from "random" and "matrix". It may be worth questioning why you are comfortable with the idea of a "random vector", when you could instead just use an array of random values without any implications of linear algebra being meaningful. | |
| S Nov 4, 2016 at 11:48 | history | suggested | psmears | CC BY-SA 3.0 |
Improve grammar and wording
|
| Nov 4, 2016 at 10:59 | review | Suggested edits | |||
| S Nov 4, 2016 at 11:48 | |||||
| Nov 4, 2016 at 9:06 | history | edited | kjetil b halvorsen♦ |
edited tags
|
|
| Nov 4, 2016 at 7:17 | answer | added | Eric Towers | timeline score: 5 | |
| Nov 4, 2016 at 2:09 | history | edited | Jeremy Miles | CC BY-SA 3.0 |
fixed typo
|
| Nov 3, 2016 at 22:55 | comment | added | Has QUIT--Anony-Mousse | Random matrixes are just a special case of random tensors. | |
| Nov 3, 2016 at 22:03 | answer | added | bright-star | timeline score: 4 | |
| Nov 3, 2016 at 20:57 | comment | added | seanv507 | @Aksakal I think the OP's point is when is it useful to analyse something as random matrices. eg in image classification you typically turn your image matrices into vectors..there is no matrix 'analysis'. so whuber's comment is the best answer so far: eg a covariance matrix has to be positive semi definite - if you want to simulate random covariance matrices its easier to work with a matrix specification than a vector. | |
| Nov 3, 2016 at 20:35 | comment | added | Aksakal | to add to @whuber : in some programming languages everything's a matrix. a scalar is a 1x1 matrix, so a random number is actually a random 1x1 matrix | |
| Nov 3, 2016 at 20:28 | answer | added | Clusterfari | timeline score: 8 | |
| Nov 3, 2016 at 20:15 | answer | added | Alex R. | timeline score: 23 | |
| Nov 3, 2016 at 19:07 | comment | added | whuber♦ | You might as well ask why matrices are of any interest. It is perfectly natural to view as random any matrix used to represent a phenomenon observed or measured in the real world. This results in a plethora of possible types and models for random matrices, ranging from adjacency matrices of random graphs to sample covariance matrices and more. | |
| Nov 3, 2016 at 18:54 | answer | added | Aksakal | timeline score: 13 | |
| Nov 3, 2016 at 18:46 | comment | added | Repmat | I imagine it's for the same we reason we have matrices in math in general. It makes neatly compacted formulas, regardless of the number of variables. | |
| Nov 3, 2016 at 18:43 | history | edited | dsaxton | CC BY-SA 3.0 |
edited title
|
| Nov 3, 2016 at 18:38 | comment | added | dsaxton | Possibly relevant: en.wikipedia.org/wiki/Random_projection. | |
| Nov 3, 2016 at 18:30 | comment | added | Matthew Gunn | I think you're fine conceptually thinking about it as a random vector that has been rearranged so that it's matrix. | |
| Nov 3, 2016 at 18:24 | history | edited | Eduardo | CC BY-SA 3.0 |
added 49 characters in body; edited title
|
| Nov 3, 2016 at 18:18 | history | asked | Eduardo | CC BY-SA 3.0 |