Timeline for Linear Model Residual Bootstrapping
Current License: CC BY-SA 3.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 11, 2017 at 3:40 | vote | accept | Alex | ||
| Oct 11, 2017 at 3:40 | comment | added | Alex | Let us continue this discussion in chat. | |
| Oct 11, 2017 at 3:38 | comment | added | Frans Rodenburg |
No, use the original residuals. In section 3.2 of the document you refer to: e <- residuals(mod.duncan.hub) ... y <- fit + e[indices]
|
|
| Oct 11, 2017 at 3:37 | comment | added | Alex | If I were to iterate this over multiple bootstraps would I have to replace original.Y with boot.Y? | |
| Oct 11, 2017 at 3:32 | comment | added | Frans Rodenburg | I have updated my answer to include this method | |
| Oct 11, 2017 at 3:32 | history | edited | Frans Rodenburg | CC BY-SA 3.0 |
Included fixed-x resampling
|
| Oct 11, 2017 at 3:29 | comment | added | Frans Rodenburg | Ah, Googling 'fixed x resampling' clarified that, I was unfamiliar with this technique. Perhaps you could include it in the question. To answer your comment: Yes, you resample from the original residuals in each step. | |
| Oct 11, 2017 at 3:26 | comment | added | Alex | I am specifically trying to create a bootstrap model using fixed-x resampling. Section 3.2 of this document is what I am trying to replicate | |
| Oct 11, 2017 at 3:18 | comment | added | Frans Rodenburg | Bootstrapping means to sample with replacement from the original sample $(y_i, x_i)$, as shown in the example code. Could you clarify why you insist on sampling the residuals? | |
| Oct 11, 2017 at 3:13 | comment | added | Alex | Just to clarify, for the next iteration of bootstrap, this means I would sample with replacement from the original residuals (1.2, -1.37, -0.86, 1.03) and add them to the Y values calculated in the previous bootstrap (6.13, 4.6, ...), correct? | |
| Oct 11, 2017 at 2:04 | history | answered | Frans Rodenburg | CC BY-SA 3.0 |