Bug in MUS::MUS.calc.n.conservative (Result is wrong)

[kiwimic]

I just got to know your package, I was looking for an implementation in R MUS algorithm (especially MUS.factor (confidence.level, pct.ratio)), when I used MUS.factor (confidence.level, pct.ratio) I came across a strange result (practically always trying was 1). When I examined the function, it turned out that there are missing parentheses to divide)

library(MUS)


MUS.calc.n.conservative_VERSION_FROM_PACKAGE_CRAN_AND_GITHUB <- function(confidence.level, tolerable.error, expected.error, book.value) {
  # calculate n consevatively, as per AICPA audit guide
  pct.ratio = expected.error / tolerable.error
  conf.factor = ceiling(MUS.factor(confidence.level, pct.ratio)*100)/100
  ceiling(conf.factor / tolerable.error / book.value)
}

MUS.calc.n.conservative_VERSION_AFTER_CHANGE <- function(confidence.level, tolerable.error, expected.error, book.value) {
  # calculate n consevatively, as per AICPA audit guide
  pct.ratio = expected.error / tolerable.error
  conf.factor = ceiling(MUS.factor(confidence.level, pct.ratio)*100)/100
  ceiling(conf.factor / (tolerable.error / book.value)) ## Added brackets
}



confidence.level <- 0.95
tolerable.error <- 9 * 1000 * 1000
expected.error <- 3 * 1000 * 1000
book.value = 300 * 1000 * 1000`

## 1. Version from package ####
MUS.calc.n.conservative_VERSION_FROM_PACKAGE_CRAN_AND_GITHUB(confidence.level, tolerable.error, expected.error, book.value)
# result = 1 (BAD)
MUS::MUS.calc.n.conservative(confidence.level, tolerable.error, expected.error,book.value)
# result = 1 (BAD)
## 2. Version with updated brackets
MUS.calc.n.conservative_VERSION_AFTER_CHANGE(confidence.level, tolerable.error, expected.error, book.value)
# result = 220 (GOOD)

[gaborcsardi]

Hi, this is a read-only mirror of CRAN, please see the package authors in DESCRIPTION.

[kiwimic]

Hi, But when i click on pull request in https://github.com/alsguimaraes/MUS I'm redirected to create pull request there: master...alsguimaraes:master [image: image.png] śr., 24 paź 2018 o 10:39 Gábor Csárdi <notifications@github.com> napisał(a):

…

Hi, this is a read-only mirror of CRAN, please see the package authors in DESCRIPTION. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AQFfAn6fyPvczGCh42G0E9H79HovAPn9ks5uoCdWgaJpZM4X3ZZg> .

[gaborcsardi]

@kiwimic That's because they forked that repository from here. That is somewhat unfortunate, and does not seem like a good idea.

[kiwimic]

Ok, so there is a away to move this pull request to alsguimaraes/MUS?

[gaborcsardi]

No, I don't think so. Looks like you were editing alsguimaraes:master, so this code is already in there and no pull request is needed?

[kiwimic]

This is also my first pull request, but when I inspect cran package, and alsguimaraes/MUS on github there is still bug in code. And now when I inspect this pull request I also see that "kiwimic wants to merge 25 commits into cran:master from alsguimaraes:master" which i don't want to.. i want to just add this comment (pull request) and wanted to create single commit after