Date

5-8-2018

Document Type

Thesis

Abstract

Predictive analytics is a branch of advanced analytics that is composed of various statistical techniques where each contributes in making predictions about future scenarios and outcomes. Some of these techniques include machine learning, artificial intelligence, data mining, predictive modeling, logistic regression, etc., and the patterns found in the results can be used to identify risks and opportunity. Predictive analytics is often associated with meteorology and weather forecasting due to the fact there are many attributes to contribute to a response, but generally, it has many applications in existing growing or established businesses, especially when it comes to decision-making about revenue, customers, and productivity (Siegel, 2016). This project is focused on the banking and financing area, and I analyzed two different datasets – one in which I generated data using software, and the other from a Portuguese bank that is available to the public online. The purpose of this project was to create a list of targeted customers that are more likely to sign up for a credit card and more likely to be issued a checking account as the binary responses by using predictive analytics. I investigated the relationship between the binary response, and the predictor variables, the characteristics of customers within the dataset. Analysis procedures and logistic regression are employed which allowed me to create the most accurate model for targeting these specific group of customers.

Department

Mathematics

Thesis Comittee

Wanchunzi Yu (Thesis Advisor)

Vignon Oussa

John Pike

Copyright and Permissions

Original document was submitted as an Honors Program requirement. Copyright is held by the author.

Included in

Mathematics Commons

Share

COinS