M ileage Discount Analytics Daniel HernándezStumpfhauser PhD Lead Statistician True Mileage Inc Ryan N Morrison CEO True Mileage Inc ID: 549614
Download Presentation The PPT/PDF document "Technology & Analytics for Usage-Bas..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Technology & Analytics for Usage-Based Auto Insurance
M
ileage Discount Analytics
Daniel Hernández-Stumpfhauser PhDLead Statistician | True Mileage, Inc.
Ryan N. Morrison
CEO | True Mileage, Inc.
Slide2
2
Mileage Up
To Discount
2,50054%
5,000
39%
7,500
34%10,00026%12,50018%15,00013% 15,000 +7%
Typical on-going verified mileage discounts today:
Mileage Discount Analytics
TM
Discounts apply regardless of other rating variables.Slide3
Rating variables with the strongest mileage relationship?
- Driver Age - Driver Gender
- Urban vs. Rural - Drivers/Vehicles - Vehicle Type - Vehicle Age3
Mileage Discount Analytics TMSlide4
Rating variables with the
strongest mileage relationship
? - Driver Age - Driver Gender - Urban vs. Rural - Drivers/Vehicles - Vehicle Type - Vehicle Age
4
Mileage Discount Analytics
TMSlide5
5
S
hould a 10,000 mile vehicle get a discount?
Mileage Discount Analytics
TMSlide6
6
Not always! It would
be a double discount
for older drivers and vehicles.
Should a 10,000 mile vehicle get a discount?
Mileage Discount Analytics
TMSlide7
7
How do we resolve the
double discounting issue?Rating Mileage: The mileage a vehicle is effectively being charged for in an existing rating plan. Vehicles should only get a discount if they are below their rating mileage.
Rating Mileage function of vehicle age, driver age, and state.
Mileage Discount Analytics
TMSlide8
Rating Mileage Model
1)
Data: Unbiased national data set with hundreds of thousands of mileage observations.
2) Variables:
The most predictive rating variables are driver age and vehicle age.
3)
Goal: Estimate rating mileage, the mileage a vehicle is effectively charged for through a typical rating plan.8Slide9
Driver Age
Vehicle Age
Rating Mileage Model
Results: 2D
9Slide10
10
To
reduce double discounts risk use:
Example 1:
(
new car and mid-age driver
)
Example 2:(older car or older driver) Mileage Discount Analytics TMSlide11
Driver Age
Vehicle Age
11
Rating Mileage Table
State
adjustments to the national rating mileage table also recommended and available.
Mileage Discount Analytics
TMSlide12
Mileage Discount Analytics
TM
12Ryan N. MorrisonCEO | True Mileage, Inc.
Visit True Mileage at Booth #1
Insurance Telematics USA, Chicago, Sept 3-4, 2014
Daniel Hernández-Stumpfhauser PhD
Lead Statistician | True Mileage, Inc.