Timeline for Why does component-wise median not make sense in higher dimensions?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 10, 2020 at 20:43 | vote | accept | Alex | ||
| Apr 9, 2020 at 12:10 | comment | added | whuber♦ | @ciamej Yes, that's a good observation. The median and the marginal median enjoy useful properties, especially in relation to the $L^1$ metric and robustness. The purpose of the present thread, though, is to appreciate the sense in which the marginal median might fail to generalize some of the properties it enjoys in one dimension. | |
| Apr 9, 2020 at 11:32 | comment | added | ciamej | If you use a non-euclidean distance metric such as the taxicab (or Manhattan) metric, the presented median is optimal. | |
| Apr 8, 2020 at 21:26 | history | edited | whuber♦ | CC BY-SA 4.0 |
added 276 characters in body
|
| Apr 8, 2020 at 21:12 | history | answered | whuber♦ | CC BY-SA 4.0 |