Dissertation/Thesis Abstract

Applications of attrition models in the television industry
by Aclander, Jaime L.S., M.S., California State University, Long Beach, 2014, 69; 1526887
Abstract (Summary)

The cost of generating a new customer exceeds the cost of retaining a present customer. This belief applies to most industries where customers' loyalty became an important strategic and financial goal. In the paid television industry, viewers have many options for television services nowadays. The number of subscriptions to these services is growing day by day, becoming a multimillion-dollar business. Individuals usually subscribe to a single television provider, but they are continuously exposed to tempting offers from the competitors. With so many alternatives, keeping the subscribers' loyalty is a challenge that can be addressed by applying statistical methods. This project analyzes 12 months of customers' data from a satellite television provider in a South American country. It calculates the monthly attrition (churn) risk for each subscriber using data associated to her/his profile and 12 months of historical call center's contacts records. The attrition risk-scores are calculated with (a) logistic regression and (b) Cox proportional hazard models. The results are compared and validated by counting the actual attrition occurring within 30 days after scoring. Preliminary processes of data analysis and discovery are discussed and implemented. The results are validated and applied within the actual operational methodology. For the top 5% of the base (with higher churn risk), these models' predictions identify over 25% of the monthly churners on average.

Indexing (document details)
Advisor: Safer, Alan
Commitee: Korosteleva, Olga, Suaray, Kagba N.
School: California State University, Long Beach
Department: Mathematics and Statistics
School Location: United States -- California
Source: MAI 53/06M(E), Masters Abstracts International
Subjects: Marketing, Applied Mathematics, Statistics
Keywords: Attrition, Customer churn, Television churn
Publication Number: 1526887
ISBN: 978-1-321-27687-9
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy