rapidsurveys / bbw Goto Github PK

Dear maintainer,

Please see the problems shown on
https://cran.r-project.org/web/checks/check_results_bbw.html.

Specifically, please see the NOTEs for the "Rd files" check.

In Rd \describe, \arguments and \value the item entries are of the form

\item{LABEL}{DESCRIPTION}

with a non-empty label: please modify your Rd files accordingly.

Please correct before 2022-05-30 to safely retain your package on CRAN.

Best,
-k

Context

Appveyor build failure after rebuilding of package site using latest pkgdown version

Currently, bbw takes a relatively long time to perform bootstrap even if the dataset is at the smallest possible number (circa 192) and at the smallest number of replicates (replicates = 399). This usually takes about 10 minutes to complete.

C++ via Rcpp might offer significant efficiencies when applied to the bbw algorithm.

• Expand current vignette

• Add vignette on step by step use of the bbw using the sample dataset

Something like...

get_estimates <- funciton(bootDF) { quantile(bootDF, probs = c(0.025, 0.5, 0.975) }