In the following lines of code,I am trying to find the % of fraudulent cases in each sub-group,like % of fraudulent cases within unmarried females as a percentage of unmarried females.Similarly for other groups like married females,married males & unmarried males.
insurance_males <- subset(insurance,gender == 1) insurance_males_unmarried <- subset(insurance_males,marital == 0) insurance_males_married <- subset(insurance_males,marital == 1) insurance_female <- subset(insurance,gender == 0) insurance_female_unmarried <- subset(insurance_female,marital == 0) insurance_female_married <- subset(insurance_female,marital == 1) table(insurance_males_unmarried$fraudulent)/nrow(insurance_males_unmarried) table(insurance_males_married$fraudulent)/nrow(insurance_males_married) table(insurance_female_unmarried$fraudulent)/nrow(insurance_female_unmarried) table(insurance_female_married$fraudulent)/nrow(insurance_female_married)
This is the output I am getting:
I wanted to know if there is shorter method to do this.I can use:
library(data.table) DT <- data.table(DF) DT[, Mean:=mean(X), by=list(Y, Z)]
to do group by on multiple groups but how do I find the proportions based on each subgroup??
Can somebody please help me with this??