I have data frame and want to calculate the AR by manual in R not by the package:
mydata <- data.frame(gender = c("Female", "Male", "Male", "Male",
"Female", "Male", "Female", "Male", "Male"),
flag = c(1,0,1,0,1,1,0,1,0))
mydata %>%
arrange(desc(gender))
I have done it with excel but i always fail to run it use R
# Cumulative % pop
random = 1:length(mydata$pred)/length(mydata$pred)
# Cumulative % of Bads
cumpercentbad = cumsum(mydata$flag)/sum(mydata$flag)
# Calculate AR
random = c(0,random)
cumpercentbad = c(0,cumpercentbad)
idx = 2:length(cumpercentbad)
testdf=data.frame(cumpercentpop = (random[idx] - random[idx-1]),
cumpercentbad = (cumpercentbad[idx] + cumpercentbad[idx-1]))
Area = sum(testdf$cumpercentbad * testdf$cumpercentpop/2)
Numerator = Area - 0.5
p(Bad) = sum(mydata$flag)/length(mydata$flag)
Denominator = 0.5*(1-Bad)
(AR = Numerator / Denominator)
The result from this 20% and I try another one variable, it gives different result