Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting
| printedLines <- utils::capture.output(split = TRUE, { | ||
| if (verbose) printf("Bootstrap %d/%d...\n", i, R) | ||
|
|
||
| sample_i <- population[sample(P, N), ] | ||
| sample_i <- resample(population, cluster = cluster) | ||
| argList_i <- c(argList, list(data = sample_i)) |
There was a problem hiding this comment.
Replacing population[sample(P, N), ] with resample(population, cluster = cluster) causes parametric bootstraps to draw P rows per replicate, where P is the simulated population size (min(P.max, N * R)). resample() resamples NROW(population) rows, so parametric runs now fit each bootstrap replicate on the entire simulated population rather than the original sample size N, drastically inflating computation and producing statistics for an incorrect sample size. The resampling helper should only take N draws from the simulated population in the parametric case.
Useful? React with 👍 / 👎.
1 participant
No description provided.