Timeline for Estimating Probability Density for Sample
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 23 at 1:29 | comment | added | Glen_b | If your problem is related to extreme values, it might be worth looking into extreme value theory. There are limit theorems for maxima/minima (leading to GEV distributions) and for asymptotic tail distributions | |
| Jun 14 at 0:23 | comment | added | Glen_b | You might be able to make some assumptions about the form of the extreme tail based on knowledge of the process and some suitable abstraction of its key features, for example. | |
| Jun 12 at 14:13 | comment | added | Glen_b | 1. In my last few paragraphs I explain the difficulty (there's no reason to think any specific curve that fits where you have data will also fit where you don't have data), and in my final paragraph I explain some options for bringing in some information/assumptions where you have no data in the sample. 2. You don't clearly explain why you need to plot "the distribution of all possible outcomes". Clarifying that might perhaps lead somewhere | |
| Jun 12 at 11:35 | comment | added | Ahmed Jyad | Thanks for the response. I now understand that using an arbitrary list of distributions does not make sense from your answer, since all I'm doing here is trying to fit observations into a distribution with no rhyme or reason. What do you suggest would be the right approach here? | |
| Jun 9 at 0:04 | history | edited | Glen_b | CC BY-SA 4.0 |
added 303 characters in body
|
| Jun 8 at 6:10 | history | edited | Glen_b | CC BY-SA 4.0 |
added 1287 characters in body
|
| Jun 8 at 5:57 | history | answered | Glen_b | CC BY-SA 4.0 |