CAS Predictive Modeling Seminar. Predictive Modeling for a Commercial Insurance Company with Little or No Data Scott Bronstein October 5, 2004. Discussion points. Data available for predictive modeling Models in use Sample results Summary. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Modeling guidelines with minimal dataModeling guidelines with minimal data Know your target populationKnow your target population Tap into business data - even minimal amount Tap into business data - even minimal amount
of business performance data can be predictiveof business performance data can be predictive Use consumer data as appropriateUse consumer data as appropriate Segment the population Segment the population Validate at logical intervals - can recalibrate if Validate at logical intervals - can recalibrate if
Intelliscore overviewIntelliscore overview Used primarily in small business lending, commercial Used primarily in small business lending, commercial
card, leasing, telecommunications, and business card, leasing, telecommunications, and business servicesservices
Commercial IntelliscoreCommercial Intelliscore Utilizes Utilizes commercial credit, business demographics, credit, business demographics,
public record and legal informationpublic record and legal information Small Business IntelliscoreSmall Business Intelliscore
Utilizes Utilizes commercial and consumer credit, business credit, business demographics, public record and demographics, public record and legal informationlegal information
Intelliscore models are Intelliscore models are segmented scoring systemssegmented scoring systems
Businesses within a modeling Businesses within a modeling sample are clustered or sample are clustered or segmented by common segmented by common characteristics such ascharacteristics such as
Size of businessSize of business Credit historyCredit history Data type availabilityData type availability
When predictors behave When predictors behave differently between clusters, the differently between clusters, the file should be segmented so the file should be segmented so the clusters can be modeled clusters can be modeled independently to capture subtle independently to capture subtle nuances in payment behaviornuances in payment behavior
Model segment integration Model segment integration Scoring system integrates model segments into one Scoring system integrates model segments into one uniform scoreuniform score
Each Intelliscore model solution consists of several Each Intelliscore model solution consists of several modeled segments each with its own set of variables modeled segments each with its own set of variables and raw scoreand raw score
SummarySummary Wide array of business data is availableWide array of business data is available Familiarity with how to use the data is Familiarity with how to use the data is
criticalcritical Small business is broadly defined and data Small business is broadly defined and data
availability will varyavailability will vary Segment the populationSegment the population Commercial risk scores are predictive Commercial risk scores are predictive
across several industries - and probably across several industries - and probably othersothers