Page 224 - Contributed Paper Session (CPS) - Volume 1

CPS1283 Kelvin H.C.Y et al.
The Generalized Linear Model output is relatively easy to explain to
stakeholders. The riskiness of a specific rating factor’s characteristics is
expressed through the fitted and computed parameters.
Based on the hypothetical examples above
• The lower relativity factor of 0.975 for females compared to
males implies that, all else being equal, females are 2.5% less
risky compared to males. Hence a 2.5% discount is justified for
females relative to males.
• The higher relativity factor of 1.250 for young drivers below 25
years old compared to drivers between 36 to 40 years old
implies that, all else being equal, young drivers below 25 years
old are 25% more risky compared to drivers between 36 to 40
years old. Hence a 25% loading is justified for young drivers
below 25 years old relative to drivers between 26 to 40 years
old.
Fire
A sample hypothetical output of the Generalized Linear Model is shown
below:
Base Premium: $ 2,500

Rating Factor 1 – Construction Class
Construction Class 1A 1.000
Construction Class 1B 1.050
Construction Class 2 1.250
Construction Class 3 2.500

Rating Factor 2 – Type of properties
Residential 0.450
Retail trading 2.000
Hotel/Offices 1.000
Food processing industries 0.950
Textiles 3.400
Timber 5.250

Using the above sample hypothetical output, the computation of Fire
insurance premium for three risk profile examples are shown below:

Building 1’s risk profile: Construction Class 1A, Residential
Fire insurance premium = $2,500 x 1.000 x 0.450 = $1,125.00

Building 2’s risk profile: Construction Class 3, Timber
Fire insurance premium = $2,500 x 2.500 x 5.250 = $32,812.50

Building 3’s risk profile: Construction Class 1B, Hotel
Fire insurance premium = $2,500 x 1.050 x 1.000 = $2,627.10

213 | I S I W S C 2 0 1 9