Timeline for Benefit of parametric bootstrap over nonparametric bootstrap
Current License: CC BY-SA 4.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 6, 2021 at 18:30 | history | edited | BruceET | CC BY-SA 4.0 |
TeX
|
| Dec 6, 2021 at 18:22 | history | edited | BruceET | CC BY-SA 4.0 |
TeX
|
| Dec 6, 2021 at 15:16 | vote | accept | John Smith | ||
| Dec 4, 2021 at 7:00 | comment | added | BruceET |
Please read sections of my Answ on parametric bootstrap again. I have modified the description, trying to make the role of the observed sample mean (a,obs in code) more clear.
|
|
| Dec 4, 2021 at 6:57 | history | edited | BruceET | CC BY-SA 4.0 |
added 98 characters in body
|
| Dec 4, 2021 at 6:48 | history | edited | BruceET | CC BY-SA 4.0 |
added 98 characters in body
|
| Dec 4, 2021 at 6:34 | comment | added | BruceET | It is possible to know (or to be reasonably sure) that the dist'n of the population is normal, exponential, gamma, beta, etc. without knowing the parameters. Then suppose you can estimate the parameters of that distribution. If so, you can use the parametric bootstrap. Estimate the parameters and use them to est, the specific dist'n from which to re-sample for a parametric bootstrap. In general, parametric bootstrap CIs will be better than if you use a parametric bootstrap where your info is restricted to the sample itself. // My param. bootstrap CIs above are narrower than nonparam. ones. | |
| Dec 4, 2021 at 4:30 | comment | added | John Smith | For clarification sake, I mean the parametric bootstrap. I guess I'm trying to rationalize when to use the parametric bootstrap since in reality we never really know the distribution of the data. So would it always be safer to default to the nonparametric. | |
| Dec 4, 2021 at 4:21 | comment | added | John Smith | Hi Bruce, very nice answer. Thanks. If the parametric assumption is wrong, can things go really bad? | |
| Dec 4, 2021 at 2:22 | history | edited | BruceET | CC BY-SA 4.0 |
edited body
|
| Dec 4, 2021 at 2:13 | history | edited | BruceET | CC BY-SA 4.0 |
edited body
|
| Dec 4, 2021 at 1:59 | history | edited | BruceET | CC BY-SA 4.0 |
edited body
|
| Dec 4, 2021 at 1:35 | history | answered | BruceET | CC BY-SA 4.0 |