PerformDailyMonetisationTests: Compares treatment to control on ARPDAU, ARPPU and Conversion...

In rvladimiro/dafR: Data Analysis Framework in R

Description

Receives a dataset with a response column. It is expected for this dataset to contain one row per DAU, including non-paying users. The function provides one of three options for measures to test and construct confidence intervals: arpdau: this will use the entirety of the dataset arppu: this will subset the datset to users whose revenue > 0 conversion: this will create a biinary column, on which 1 identifies a user who did a purchase

This function should also receive a character vector that identifies the name of the control group, as well as the name of the column that contains the revenue for each DAU

This functions combines the functions PerformPermutation and GetBootstrappedCI to test the most common monetisation KPIs against all avaliable treatement groups

Returns all the p-values and confidence interval boundaries for each treatement. It also returns the calculated statistics for each bootstrap iteration

Arguments

dataset	The dataset containing one observation per row
kpi	One of the following: 'arpdau', 'arppu', 'conversion'
revenueColumn	The name of the column that contains the measures on revenue per observation
groupColumn	The name of the column that contains the group identification
controlName	The name which identifies the control group in the dataset
bootIterations	The number of bootstrap resamples to perform
bootConfLevel	The confidence level for the interval. Between 1 and 100
...	Additional arguments to pass on to permTS function

Value

List containing two dataframes. The first dataset contains all the p-values and CI boundaries. The second dataframe contains all calculated estimates for the mean of each group

Author(s)

João Monteiro

rvladimiro/dafR documentation built on June 26, 2019, 4:37 a.m.