Timeline for Could fitting a Poisson distribution be better than just counting random events?
Current License: CC BY-SA 4.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 5, 2020 at 18:44 | comment | added | NotMe | This depends on what you like to get. If you take a "simple" Poissionian distribution, every serious statistical program will provide a confidence interval to the fit parameter. However, if you prefer a prediction interval you probable need to use some extra steps. And if you consider multiple decays I expect that it gets messy pretty fast. | |
| Jan 5, 2020 at 17:39 | comment | added | KF Gauss | Since it is just simple Poisson statistics, I'm guessing there is an exact expression for the uncertainties in either case (mean count rate vs fitting). Or aft the very least it should be easily simulated. | |
| Jan 5, 2020 at 16:35 | history | answered | NotMe | CC BY-SA 4.0 |