Timeline for What is the problem with $p > n$?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 6, 2019 at 21:56 | comment | added | boomkin | You can call it a rare cancer disease, but in my experience it takes time to gather enough data even for more common ones if you are limited to one hospital’s patients. The key is that n is scarce, while p is not. I’m still looking for a good, practical example, but David Wipf’s work on neuroimaging by overcomplete dictionaries might be a good start from the applied theoretical perspective. | |
| Jan 5, 2019 at 13:53 | comment | added | EconBoy | Thank you a lot for the answer, I would be grateful if you give me a case for study. From what I heard from you it sounds like "the case where it is rare and the researcher really want to dig deep into that thing with the huge amount of variable they think they will extract the useful pattern and information to predict that rarity?" | |
| Jan 5, 2019 at 13:46 | vote | accept | EconBoy | ||
| Jan 5, 2019 at 13:30 | history | answered | boomkin | CC BY-SA 4.0 |