ACCOUNT: Predicting whether a customer will open a new kind of account

In regclass: Tools for an Introductory Class in Regression and Modeling

Description

Customers were marketed a new type of account at a bank. It is desired to model what factors seemed to be associated with the probability of opening the account to tune marketing strategy.

Usage

data("ACCOUNT")

Format

A data frame with 24242 observations on the following 8 variables.

Purchase

a factor with levels No Yes

Tenure

a numeric vector, the number of years the customer has been with the bank

CheckingBalance

a numeric vector, amount currently held in checking (may be negative if overdrafted)

SavingBalance

a numeric vector, amount currently held in savings (0 or larger)

Income

a numeric vector, yearly income in thousands of dollars

Homeowner

a factor with levels No Yes

Age

a numeric vector

Area.Classification

a factor with levels R S U for rural, suburban, or urban

Details

Who is more likely to open a new type of account that a bank wants to try to sell its customers? Try logistic regression or partition models to see if you can develop a model that accurately classifies purchasers vs. non-purchasers. Or, try to develop a model that does well in promoting to nearly all customers who would buy the account.

regclass documentation built on March 26, 2020, 8:02 p.m.