Please use this identifier to cite or link to this item: http://hdl.handle.net/11718/14032
Title: Predicting Policy Renewal Probability in Auto Insurance Sector
Authors: Pagidimarri, Venkatesh
Kasivajjala, Vamsichandra
Miyan, Deepti
Keywords: Predictive models;Insurance;Policy;Renewal Rate;Logistic regression;Segmentation;Business Impact
Issue Date: 2015
Publisher: Indian Institute of Management, Ahmedabad
Citation: Pagidimarri, V., Kasivajjala, V., & Miyan, D.. (2015). Predicting Policy Renewal Probability in Auto Insurance Sector. 4th IIMA International Conference on Advanced Data Analysis, Business Analytics and Intelligence. Indian Institute of Management, Ahmedabad
Series/Report no.: IC 15;032
Abstract: We are entering a new phase of the information era, in which organizations query and analyze huge volumes of diverse data in real-time to improve outcomes for their most critical business processes. In today’s competitive edge, the concept of Predictive Analytics is used in every business sector. This paper discusses about how Predictive Analytics helps in increasing the policy renewal rate for Auto Insurance sector. Customer demographics data, Type of vehicle, historical claims data is used to understand the behaviour of customers in renewing the policies. The model identifies the parameters which impacts the policy renewal rates in customers. Customers were grouped in to High, Medium and Low segments based on their policy renewal probabilities. This will help the insurers to identify their potential customers that are more likely to go for the policy renewal. The last section of the paper outlines the financial benefits in increasing the policy renewal rate using the predictive model.
URI: http://hdl.handle.net/11718/14032
Appears in Collections:4th IIMA International Conference on Advanced Data Analysis, Business Analytics and Intelligence

Files in This Item:
File Description SizeFormat 
IC 15-032.pdf
  Restricted Access
383.6 kBAdobe PDFView/Open Request a copy


Items in IIMA Institutional Repository are protected by copyright, with all rights reserved, unless otherwise indicated.