Skip to content

Instantly share code, notes, and snippets.

@fauxneticien

fauxneticien/cis-by-condition.r

Last active December 7, 2021 23:52
Show Gist options
  • Save fauxneticien/391f1b8a593e29bbd7c6b2622f7bea4e to your computer and use it in GitHub Desktop.
Save fauxneticien/391f1b8a593e29bbd7c6b2622f7bea4e to your computer and use it in GitHub Desktop.
Download ZIP
Parametric bootstrap estimates for A/B test conditions
Raw
cis-by-condition.r
library(boot)
library(purrr)
n_bootstraps <- 100
sample_data <- read.csv("~/Desktop/sample-data.csv", stringsAsFactors = FALSE)
get_mean_pdiff <- function(data, indices) {
d <- data[indices,]
return(mean(d$prop_diff))
}
cis_by_condition <- imap_dfr(
.x = split(sample_data, sample_data$condition),
.f = function(data, condition) {
results <- boot(data=data, statistic = get_mean_pdiff, R = n_bootstraps, parallel = "multicore")
conf_is <- boot.ci(results, type="bca")
data.frame(
condition = condition,
prop_diff = conf_is$t0,
ci_lower = conf_is$bca[4],
ci_upper = conf_is$bca[5]
)
}
)
Raw
sample-data.csv
subject condition oct_total_spent oct_hrm_spent oct_spent_prop nov_total_spent oct_hrm_spent nov_spent_prop prop_diff
C01 control 100 80 0.8 100 75 0.75 -0.05
C02 control 120 100 0.833333333 120 110 0.916666667 0.083333333
C03 control 90 76 0.844444444 90 70 0.777777778 -0.066666667
C04 control 75 50 0.666666667 75 55 0.733333333 0.066666667
I01 intervention 114 100 0.877192982 114 90 0.789473684 -0.087719298
I02 intervention 123 100 0.81300813 123 85 0.691056911 -0.12195122
I03 intervention 192 150 0.78125 192 100 0.520833333 -0.260416667
I04 intervention 210 150 0.714285714 210 160 0.761904762 0.047619048
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment