Fortis Bank Insurance

Corporate Actuarial Control

The chain ladder method

Calculation of the proportionality factors.

The cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

The d run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 1.362 1.009 1.004 1.002 1.005
1 1.384 1.008 1.005 1.002
2 1.381 1.010 1.004
3 1.396 1.017
4 1.378
5
6

#DIV/0! #DIV/0! 0.9957 0.9981 0.9953

pj 1.00434 1.00186 1.00474 DEVELOPMENT FACTORS
98.914% 98.914% 98.914% 99.344% 99.529% DEVELOPMENT RATIOS
98.914% 0.000% 0.000% 0.430% 0.185% DEVELOPMENT PERCENTAG

Tot. Loss 24,832 27,087 20,525 20,614 14,595 14,664 14,664

169% 185% 140% 141% 100% 100% 100%
Tol. Increm. 24,832 2,255 (6,562) 89 38 98 -
169% 15% -45% 1% 0% 1% 0%

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Completion of the cumulative run-off triangle

Method 1 (using the proportionality factors p i,i+1)
Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372.00 4,411.00 4,428.00 4,435.00 4,456.00 4,456.00
1 3,367 4,659.00 4,696.00 4,720.00 4,730.00 4,752.40 4,752.40

Loss Reserving Techniques in Non-Life 1 The chain ladder method

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

2 3,871 5,345.00 5,398.00 5,420.00 5,430.07 5,455.78 5,455.78

3 4,239 5,917.00 6,020.00 6,046.15 0.00 0.00
4 4,929 6,794.00 0.00 0.00 0.00 0.00 0.00
5 5,217 0.00 0.00 0.00 0.00
6

Method 2 (using the proportionality factors p 1,i)

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456 4,456.00
1 3,367 4,659 4,696 4,720 4,730 4,752.40 4,752.40
2 3,871 5,345 5,398 5,420 5,430.07 5,455.78 5,455.78
3 4,239 5,917 6,020 6,046.15 6,057.38 6,086.06 6,086.06
4 4,929 6,794 0.00 0.00 0.00 0.00 0.00
5 5,217 0.00 0.00 0.00 0.00 0.00 0.00
6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21 -
1 3,367 1,292 37 24 10 22.40 -
2 3,871 1,474 53 22 10.07 25.71 -
3 4,239 1,678 103 26.15 11.24 28.68 -
4 4,929 1,865 ### - - - -
5 5,217 (5,217.00) - - - - -
6

Estimated Reserves
Year of origin Reserve Reserve
0 0 0
1 22 22
2 36 36
3 66
4 (6,794.00)
5 -5,217 -5,217
Total:

The d-triangle

Loss Reserving Techniques in Non-Life 2 The chain ladder method

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Development year
Year of origin 0 1 2 3 4 5 6
0 1.362 1.009 1.004 1.002 1.005 1.000
1 1.384 1.008 1.005 1.002 1.005 1.000
2 1.381 1.010 1.004 1.002 1.005 1.000
3 1.396 1.017 1.004 - #DIV/0! #DIV/0!
4 1.378 - #DIV/0! #DIV/0! #DIV/0! #DIV/0!
5 - #DIV/0! #DIV/0! #DIV/0! #DIV/0! #DIV/0!
6

Loss Reserving Techniques in Non-Life 3 The chain ladder method

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

DEVELOPMENT FACTORS
DEVELOPMENT RATIOS
DEVELOPMENT PERCENTAGES

141.5%

Loss Reserving Techniques in Non-Life 4 The chain ladder method

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The (old) chain ladder method

Here the proportionality factors are the mean of empirical proportionality factors.

Calculation of the proportionality factors.

The cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

7,000 6,250
6,050
f(x) = 1.4039538088x
1.3814972301x - 90.3107918789
6,500 5,850 f(x) = 1.0404611982x
1.011835442x - 147.2697736397
5,650
6,000
5,450
5,500 5,250
5,050
5,000 4,850
4,650
4,500
4,450
4,000 4,250

3,000 3,500 4,000 4,500 5,000 5,500 4,000 4,500 5,000 5,500 6,000 6,500

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 1.362 1.009 1.004 1.002 1.005
1 1.384 1.008 1.005 1.002
2 1.381 1.010 1.004
3 1.396 1.017
4 1.378
5
6

pj 1.00435 1.00185 1.00474

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Completion of the cumulative run-off triangle

Loss Reserving Techniques in Non-Life 5 The chain ladder method (2)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Method 1 (using the proportionality factors p i,i+1)

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372.00 4,411.00 4,428.00 4,435.00 4,456.00 4,456.00
1 3,367 4,659.00 4,696.00 4,720.00 4,730.00 4,752.40 4,752.40
2 3,871 5,345.00 5,398.00 5,420.00 5,430.03 5,455.74 5,455.74
3 4,239 5,917.00 6,020.00 6,046.17 6,057.35 6,086.03 6,086.03
4 4,929 6,794.00 0.00 0.00 0.00 0.00
5 5,217 0.00 0.00 0.00 0.00 0.00 0.00
6

Method 2 (using the proportionality factors p 1,i)

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456 4,456.00
1 3,367 4,659 4,696 4,720 4,730 4,752.40 4,752.40
2 3,871 5,345 5,398 5,420 5,430.03 5,455.74 5,455.74
3 4,239 5,917 6,020 6,046.17 6,057.35 6,086.03 6,086.03
4 4,929 6,794 0.00 0.00 0.00 0.00 0.00
5 5,217 0.00 0.00 0.00 0.00 0.00 0.00
6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21 -
1 3,367 1,292 37 24 10 22.40 -
2 3,871 1,474 53 22 10.03 25.71 -
3 4,239 1,678 103 26.17 11.18 28.68 -
4 4,929 1,865 ### - - - -
5 5,217 (5,217.00) - - - - -
6

Estimated Reserves
Year of origin Reserve Reserve
0 0 0
1 22 22
2 36 36
3 66 66
4 -6,794
5 -5,217 -5,217
Total: -11,886.8 (Official Chain Ladder : )

The d-triangle

Loss Reserving Techniques in Non-Life 6 The chain ladder method (2)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Development year
Year of origin 0 1 2 3 4 5 6
0 1.362 1.009 1.004 1.002 1.005 1.000
1 1.384 1.008 1.005 1.002 1.005 1.000
2 1.381 1.010 1.004 1.002 1.005 1.000
3 1.396 1.017 1.004 1.002 1.005 1.000
4 1.378 - #DIV/0! #DIV/0! #DIV/0! #DIV/0!
5 - #DIV/0! #DIV/0! #DIV/0! #DIV/0! #DIV/0!
6

Loss Reserving Techniques in Non-Life 7 The chain ladder method (2)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Loss Reserving Techniques in Non-Life 8 The chain ladder method (2)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The (old) chain ladder method

Here the proportionality factors are the mean of empirical proportionality factors.

Calculation of the proportionality factors.

The cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

1.38093 1.01143 1.00434 1.00186 1.00474

67,0,250

6,05
f(x)=1.403f(x)=1.38 958-749720
,506 f(x)=1.4 982-73
5,80 f(x)=1. 342

Completion of the cumulative run-off triangle

5,60
6,0

5,40

55,0,20

5,0

,05
4,850 Development year
4,650
4,50

4,50
Year of origin 0 1 2 3 4 5 6
,25044,0 0 3,209 4,372.00 4,411.00 4,428.00 4,435.00 4,456.00 4,456.00
8,05,0
,04
3,0 1 3,367 4,659.00 4,696.00 4,720.00 4,730.00 4,752.40 4,752.40
2 3,871 5,345.00 5,398.00 5,420.00 5,430.07 5,455.78 5,455.78
3 4,239 5,917.00 6,020.00 6,046.15 6,057.38 6,086.06 6,086.06
4 4,929 6,794.00 6,871.67 6,901.52 6,914.34 6,947.08 6,947.08
5 5,217 7,204.33 7,286.69 7,318.34 7,331.94 7,366.66 7,366.66
6

0 0.004713 0.006559 0.01085507 0.022036

4591 4672 4863 5175 5673
0 22.01795 31.89587 56.1749909 125.0083
Reserves 0.0 24.6 35.6 62.7 139.6

Loss Reserving Techniques in Non-Life 9 cape code

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

PREMIUM

4591 4,456.00
4672 4,752.40
4863 5,455.78
5175 6,086.06
5673 6,947.08
6431 7,366.66

LOSS RATIO 35,063.99

31405
111.65%

p PREMIUM
0 4,591 0.0 0
0.004713 4,672 22.0 25
0.006559 4,863 31.9 36
0.010855 5,175 56.2 63
0.022036 5,673 125.0 140
0.291809 6,431 1876.6 2095

2358
0.291809
6431
1876.623
2095.3

Loss Reserving Techniques in Non-Life 10 cape code

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method

Calculation of the proportionality factors.

The cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

Development year
Year of origin 0 1 2 3 4 5 6
0 1.362 1.009 1.004 1.002 1.005
1 1.373 1.008 1.005 1.002
2 1.376 1.009 1.004
3 1.382 1.011
4 1.381
5
6

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435
1 3,367 4,659 4,696 4,720 4,727
2 3,871 5,345 5,398 5,422
3 4,239 5,917 5,970
4 4,929 6,811
5
6

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428
1 3,367 4,659 4,696 4,714
2 3,871 5,345 5,390
3 4,239 5,833
4
5
6

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411
1 3,367 4,659 4,701
2 3,871 5,316
3
4
5

Loss Reserving Techniques in Non-Life 11 Boni-Mali (chain ladder)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372
1 3,367 4,587
2
3
4
5
6

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 #REF! #REF! #REF! #REF! #REF! #REF!
1 3,367 #REF! #REF! #REF! #REF! #REF! #REF!
2 3,871 #REF! #REF! #REF! #REF! #REF! #REF!
3 4,239 #REF! #REF! #REF! #REF! #REF! #REF!
4 4,929 #REF! #REF! #REF! #REF! #REF! #REF!

Previous estimation (without any correction for year 0)

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 #REF! #REF! #REF! #REF! #REF! #REF!
1 3,367 #REF! #REF! #REF! #REF! #REF! #REF!
2 3,871 #REF! #REF! #REF! #REF! #REF! #REF!
3 4,239 #REF! #REF! #REF! #REF! #REF! #REF!
4 4,929 6,811 6,871.97 6,902.91 6,913.82 6,913.82 6,913.82

Previous estimation (without correction for year 0) 0.5%

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 #REF! #REF! #REF! #REF! #REF! #REF!
1 3,367 #REF! #REF! #REF! #REF! #REF! #REF!
2 3,871 #REF! #REF! #REF! #REF! #REF! #REF!
3 4,239 #REF! #REF! #REF! #REF! #REF! #REF!
4 4,929 6,811 6,871.97 6,902.91 6,913.82 6,948.39 6,948.39

Loss Reserving Techniques in Non-Life 12 Boni-Mali (chain ladder)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Development year
Year of ori 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Development year
Year of ori 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7
1 3,367 1,292 37 24 7
2 3,871 1,474 53 24
3 4,239 1,678 53 -
4 4,929 1,882 - -
5
6

Development year
Year of ori 0 1 2 3 4 5 6
0 3,209 1,163 39 17
1 3,367 1,292 37 18
2 3,871 1,474 45
3 4,239 1,594
4
5
6

Development year
Year of ori 0 1 2 3 4 5 6
0 3,209 1,163 39
1 3,367 1,292 42
2 3,871 1,445
3
4
5

Loss Reserving Techniques in Non-Life 13 Boni-Mali (chain ladder)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Development year
Year of ori 0 1 2 3 4 5 6
0 3,209 1,163
1 3,367 1,220
2
3
4
5
6

#REF!
#REF!
#REF! #REF!

#REF!
#REF!
#REF! #REF!

Loss Reserving Techniques in Non-Life 14 Boni-Mali (chain ladder)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Difference

3
(2)
50
187
237

Difference

6
8
204
218

Difference

(5)
182
177

Difference

72

Loss Reserving Techniques in Non-Life 15 Boni-Mali (chain ladder)

(E. Van den Borre)
 Formation Caritat, 2008

The chain ladder method, Average cost

Number of claims

Development year
Year of origin 0 1 2 3 4 5 6
0 1,043.4 1,045.5 1,047.5 1,047.7 1,047.7 1,047.7
1 1,043.0 1,027.1 1,028.7 1,028.9 1,028.7
2 965.1 967.9 967.8 970.1
3 977.0 984.7 986.8
4 1,099.0 1,118.5
5 1,076.3
6

1.00316 1.00139 1.00089 0.99990 1.00000

Development year
Year of origin 0 1 2 3 4 5 6
0 1,043 1,045.50 1,047.50 1,047.70 1,047.70 1,047.70 1,047.70
1 1,043 1,027.10 1,028.70 1,028.90 1,028.70 1,028.70 1,028.70
2 965 967.90 967.80 970.10 970.01 970.01 970.01
3 977 984.70 986.80 987.68 987.58 987.58 987.58
4 1,099 1,118.50 1,120.06 1,121.05 1,120.94 1,120.94 1,120.94
5 1,076 1,079.70 1,081.20 1,082.16 1,082.06 1,082.06 1,082.06
6

Paid losses

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

1.38093 1.01143 1.00434 1.00186 1.00474

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372.00 4,411.00 4,428.00 4,435.00 4,456.00 4,456.00
1 3,367 4,659.00 4,696.00 4,720.00 4,730.00 4,752.40 4,752.40

Loss Reserving Techniques in Non-Life 16 Chain ladder, average cost

([email protected])
 Formation Caritat, 2008

2 3,871 5,345.00 5,398.00 5,420.00 5,430.07 5,455.78 5,455.78

3 4,239 5,917.00 6,020.00 6,046.15 6,057.38 6,086.06 6,086.06
4 4,929 6,794.00 6,871.67 6,901.52 6,914.34 6,947.08 6,947.08
5 5,217 7,204.33 7,286.69 7,318.34 7,331.94 7,366.66 7,366.66
6

Incurred losses

Development year
Year of origin 0 1 2 3 4 5 6
0 4,975 4,629 4,497 4,470 4,456 4,456
1 5,135 4,949 4,783 4,760 4,750
2 5,681 5,631 5,492 5,470
3 6,272 6,198 6,131
4 7,326 7,087
5 7,353
6

0.96955 0.97646 0.99513 0.99740 1.00000

Development year
Year of origin 0 1 2 3 4 5 6
0 4,975 4,629.00 4,497.00 4,470.00 4,456.00 4,456.00 4,456.00
1 5,135 4,949.00 4,783.00 4,760.00 4,750.00 4,750.00 4,750.00
2 5,681 5,631.00 5,492.00 5,470.00 5,455.78 5,455.78 5,455.78
3 6,272 6,198.00 6,131.00 6,101.12 6,085.25 6,085.25 6,085.25
4 7,326 7,087.00 6,920.15 6,886.42 6,868.51 6,868.51 6,868.51
5 7,353 7,129.07 6,961.23 6,927.30 6,909.29 6,909.29 6,909.29
6

NUMBER AV. COST (I) AV. COST (TOT. COSTTOT. COSTLATEST RES (I)
0 1,047.70 4.25 4.25 4456 4456 4,456 -
1 1,028.70 4.62 4.62 4750 4752 4,730 20
2 970.01 5.62 5.62 5456 5456 5,420 36
3 987.58 6.16 6.16 6085 6085 6,020 65
4 1,120.94 6.13 6.20 6869 6947 6,794 75
5 1,082.06 6.38 6.80 6908 7360 5,217 1,691

Total 1,886

Loss Reserving Techniques in Non-Life 17 Chain ladder, average cost

([email protected])
 Formation Caritat, 2008

AVERAGE Paid losses

Development year
Year of ori 0 1 2 3 4 5 6
0 3.08 4.18 4.21 4.23 4.23 4.25
1 3.23 4.54 4.56 4.59 4.60
2 4.01 5.52 5.58 5.59
3 4.34 6.01 6.10
4 4.48 6.07
5 4.85
6

1.37541 1.01013 1.00330 1.00196 1.00474

Development year
Year of ori 0 1 2 3 4 5 6
0 3.08 4.18 4.21 4.23 4.23 4.25 4.25
1 3.23 4.54 4.56 4.59 4.60 4.62 4.62

Loss Reserving Techniques in Non-Life 18 Chain ladder, average cost

([email protected])
 Formation Caritat, 2008

2 4.01 5.52 5.58 5.59 5.60 5.62 5.62

3 4.34 6.01 6.10 6.12 6.13 6.16 6.16
4 4.48 6.07 6.14 6.16 6.17 6.20 6.20
5 4.85 6.67 6.73 6.76 6.77 6.80 6.80
6

AVERAGE Incurred losses

Development year
Year of ori 0 1 2 3 4 5 6
0 4.77 4.43 4.29 4.27 4.25 4.25
1 4.92 4.82 4.65 4.63 4.62
2 5.89 5.82 5.67 5.64
3 6.42 6.29 6.21
4 6.67 6.34
5 6.83
6

0.96618 0.97530 0.99412 0.99751 1.00000

Development year
Year of ori 0 1 2 3 4 5 6
0 4.77 4.43 4.29 4.27 4.25 4.25 4.25
1 4.92 4.82 4.65 4.63 4.62 4.62 4.62
2 5.89 5.82 5.67 5.64 5.62 5.62 5.62
3 6.42 6.29 6.21 6.18 6.16 6.16 6.16
4 6.67 6.34 6.18 6.14 6.13 6.13 6.13
5 6.83 6.60 6.44 6.40 6.38 6.38 6.38
6

RES (P)
-
22
36
65
153
2,143

2,419

Loss Reserving Techniques in Non-Life 19 Chain ladder, average cost

([email protected])
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method, Variant I

Estimation of the linear least square parameters
The d-triangle
Development year
Year of origin 0 1 2 3 4 5
0 1.362 1.009 1.004 1.002 1.005
1 1.384 1.008 1.005 1.002
2 1.381 1.010 1.004
3 1.396 1.017
4 1.378
5
6

Method 1 (using Excel functions)

column 0: a0 0.0044
b0 1.3714
column 1: a1 0.0027
b1 1.0069
column 2: a2 0.0001
b2 1.0042
column 3: a3 0.0005
b3 1.0016
column 4: 1 (no least square regression needed)

Method 2 (using formulas)

column 0: n 5.0
Sxi 10.0
Syi 6.9
Sxiyi 13.8
S(xi)2 30.0
a0 0.0044
b0 1.3714
column 1: n 4.0
Sxi 6.0
Syi 4.0
Sxiyi 6.1
S(xi)2 14.0
a1 0.0027
b1 1.0069
column 2: n 3.0
Sxi 3.0
Syi 3.0
Sxiyi 3.0
S(xi)2 5.0
a2 0.0001
b2 1.0042
column 3: n 2.0

Loss Reserving Techniques in Non-Life 20 Chain ladder, Variant I

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Sxi 1.0
Syi 2.0
Sxiyi 1.0
S(xi)2 1.0
a3 0.0005
b3 1.0016
column 4: 1.0047 (no least square regression needed)

Completion of the d-triangle

Method 1 (using Excel functions)
Development year
Year of origin 0 1 2 3 4 5
0 1.3624 1.0089 1.0039 1.0016 1.0047 1.0000
1 1.3837 1.0079 1.0051 1.0021 1.0047 1.0000
2 1.3808 1.0099 1.0041 1.0027 1.0047 1.0000
3 1.3958 1.0174 1.0046 1.0032 1.0047 1.0000
4 1.3784 1.0047 1.0037 1.0047 1.0000
5 1.3934 1.0206 1.0048 1.0043 1.0047 1.0000
6

Method 2 (using formulas)

Development year
Year of origin 0 1 2 3 4 5
0 1.3624 1.0089 1.0039 1.0016 1.0047 1.0000
1 1.3837 1.0079 1.0051 1.0021 1.0047 1.0000
2 1.3808 1.0099 1.0041 1.0027 1.0047 1.0000
3 1.3958 1.0174 1.0046 1.0032 1.0047 1.0000
4 1.3784 1.0179 1.0047 1.0037 1.0047 1.0000
5 1.3934 1.0206 1.0048 1.0043 1.0047 1.0000
6

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730 4,752
2 3,871 5,345 5,398 5,420 5,434 5,460
3 4,239 5,917 6,020 6,048 6,067 6,096
4 4,929 6,794 6,916 - -
5 5,217 7,270 7,420 7,455 7,487 7,523
6

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10 22
2 3,871 1,474 53 22 14 26
3 4,239 1,678 103 28 19 29
4 4,929 1,865 122 (6,916) - -
5 5,217 2,053 150 36 32 35
6

Estimated Reserves

Loss Reserving Techniques in Non-Life 21 Chain ladder, Variant I

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Year of origin Reserve

0 0
1 22
2 40
3 76
4 -6,794
5 2,306
Total: -4,350

Loss Reserving Techniques in Non-Life 22 Chain ladder, Variant I

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Loss Reserving Techniques in Non-Life 23 Chain ladder, Variant I

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

6
4,456
4,752
5,460
6,096
-
7,523

6
-
-
-
-
-
-

Loss Reserving Techniques in Non-Life 24 Chain ladder, Variant I

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method, Variant II

Calculation of the weighted average of each column (w ij =( i+j+1)2)

The d-triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 0.930 0.971 0.994 0.997 1.000
1 0.964 0.966 0.995 0.998
2 0.991 0.975 0.996
3 0.988 0.989
4 0.967
5
6

Calculation d0
t-j 5
j 0
i wi0 di0 wi0*di0
0 1 0.9305 0.930452
1 2 0.9638 1.927556
2 3 0.9912 2.973596
3 4 0.9882 3.952806
4 5 0.9674 4.836882
d0 0.974753
Calculation d1
t-j 5
j 1
i wi1 di1 wi1*di1
0 2 0.9715 1.942968
1 3 0.9665 2.899374
2 4 0.9753 3.901261
3 5 0.9892 4.94595
d1 0.977825
Calculation d2
t-j 4
j 2
i wi2 di2 wi2*di2
0 3 0.9940 2.981988
1 4 0.9952 3.980765
2 5 0.9960 4.979971
d2 0.995227
Calculation d3
t-j 3
j 3

Loss Reserving Techniques in Non-Life 25 Chain ladder, Variant II

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

i wi3 di3 wi3*di3

0 4 0.9969 3.987472
1 5 0.9979 4.989496
d3 0.997441
Calculation d4
t-j 2
j 4
i wi4 di4 wi1*di4
0 5 1.0000 5
d4 1

Completion of the d-triangle

d0 0.9748
d1 0.9778
d2 0.9952
d3 0.9974
d4 1.0000
d5 1.0000

Development year
Year of origin 0 1 2 3 4 5 6
0 0.930 0.971 0.994 0.997 1.000 1.000
1 0.964 0.966 0.995 0.998 1.000 1.000
2 0.991 0.975 0.996 0.997 1.000 1.000
3 0.988 0.989 - 0.997 1.000 1.000
4 0.967 0.978 - 0.997 1.000 1.000
5 0.975 0.978 0.997 1.000 1.000
6

The completed cumulative triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456 4,456
1 3,367 4,659 4,696 4,720 4,730 4,730 4,730
2 3,871 5,345 5,398 5,420 5,406 5,406 5,406
3 4,239 5,917 6,020 - - -
4 4,929 6,794 6,643 - - - -
5 5,217 5,085 4,973 - - - -
6

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21 -
1 3,366 1,292 37 24 10 - -
2 3,869 1,474 53 22 (14) - -
3 4,236 1,678 103 (6,020) - - -
4 4,925 1,865 (151) (6,643) - - -
5 5,212 (132) (113) (4,973) - - -
6

Loss Reserving Techniques in Non-Life 26 Chain ladder, Variant II

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Estimated Reserves
Year of origin Reserve
0 0
1 0
2 -14
3 -6,020
4 -6,794
5 -5,217
Total: -18,045

Loss Reserving Techniques in Non-Life 27 Chain ladder, Variant II

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method, Variant III

The logarithm of the cumulative run-off triangle.
The cumulative run-off triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

ln(Cumulative triangle)
Development year
Year of origin 0 1 2 3 4 5 6
0 8.07 8.38 8.39 8.40 8.40 8.40
1 8.12 8.45 8.45 8.46 8.46
2 8.26 8.58 8.59 8.60
3 8.35 8.69 8.70
4 8.50 8.82
5 8.56
6

Estimation of the lineair least square parameters

Method 1 (using Excel functions)
column 0: b0 0.104684923
a0 8.050189709
column 1: b1 0.112066733
a1 8.360432289
column 2: b2 0.107227495
a2 8.374896106
column 3: b3 0.1010739
a3 8.383298922
column 4: b4 0.064397587
a4 8.397282895
column 5: b5 Err:504
a5 Err:504

Estimation of the cost level adjustment factors

b0 0.1047 g0 1.110361
b1 0.1121 g1 1.118588
b2 0.1072 g2 1.113187
b3 0.1011 g3 1.106358
b4 0.0644 g4 1.066516
b5 Err:504 g5 Err:504

Loss Reserving Techniques in Non-Life 28 Chain ladder, Variant III

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Cost level adjustment of the incremental run-off triangle

The incremental run-off triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6 -

n-(i+j) (n = 6)
Development year
Year of origin 0 1 2 3 4 5 6
0 5.00 4.00 3.00 2.00 1.00 0.00
1 4.00 3.00 2.00 1.00 0.00
2 3.00 2.00 1.00 0.00
3 2.00 1.00 0.00
4 1.00 0.00
5 0.00
6

The x'-run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 5,416.14 1,820.79 53.80 20.81 7.47 Err:504
1 5,117.99 1,808.31 45.85 26.55 10.00
2 5,299.26 1,844.32 59.00 22.00
3 5,226.27 1,876.99 103.00
4 5,472.97 1,865.00
5 5,217.00
6

Completion of the x'-run-off-triangle

column 0
j 0
n-j 6
i wi,0 x'i,0 wi,0 x'i,0
0 1 5,416.14 5416.143
1 2 5,117.99 10235.98
2 3 5,299.26 15897.79
3 4 5,226.27 20905.07
4 5 5,472.97 27364.84
5 6 5,217.00 31302
6 7 0.00 0
28 111121.8
x'0 3968.63636

Loss Reserving Techniques in Non-Life 29 Chain ladder, Variant III

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

column 1
j 1
n-j 5
i wi,01 x'i,1 wi,0 x'i,1
0 2 1,820.79 3641.578
1 3 1,808.31 5424.924
2 4 1,844.32 7377.299
3 5 1,876.99 9384.949
4 6 1,865.00 11190
5 7 0.00 0
27 37018.75
x'1 1371.06482
column 2
j 2
n-j 4
i wi,2 x'i,2 wi,2 x'i,2
0 3 53.80 161.3953
1 4 45.85 183.3996
2 5 59.00 294.9947
3 6 103.00 618
4 7 0.00 0
25 1257.79
x'2 50.3115809
column 3
j 3
n-j 3
i wi,3 x'i,3 wi,3 x'i,3
0 4 20.81 83.23397
1 5 26.55 132.763
2 6 22.00 132
3 7 0.00 0
22 347.997
x'3 15.8180443
column 4
j 4
n-j 2
i wi,4 x'i,4 wi,4 x'i,4
0 5 7.47 37.32807
1 6 10.00 60
2 7 0.00 0
18 97.32807
x'4 5.40711512
column 5
j 5
n-j 1
i wi,5 x'i,5 wi,5 x'i,5
0 6 Err:504 Err:504
1 7 0.00 0
13 Err:504
x'5 Err:504

Loss Reserving Techniques in Non-Life 30 Chain ladder, Variant III

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

column 6
j 6
n-j 0
i wi,6 x'i,6 wi,6 x'i,6
0 7 0.00 0
7 0
x'6 0

The completed x'-run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 5,416.14 1,820.79 53.80 20.81 7.47 Err:504 0.00
1 5,117.99 1,808.31 45.85 26.55 10.00 Err:504 0.00
2 5,299.26 1,844.32 59.00 22.00 5.41 Err:504 0.00
3 5,226.27 1,876.99 103.00 15.82 5.41 Err:504 0.00
4 5,472.97 1,865.00 50.31 15.82 5.41 Err:504 0.00
5 5,217.00 1,371.06 50.31 15.82 5.41 Err:504 0.00
6

Reversion of the cost level adjustment

i+j-n
Development year
Year of origin 0 1 2 3 4 5 6
0 -5.00 -4.00 -3.00 -2.00 -1.00 0.00
1 -4.00 -3.00 -2.00 -1.00 0.00 1.00
2 -3.00 -2.00 -1.00 0.00 1.00 2.00
3 -2.00 -1.00 0.00 1.00 2.00 3.00
4 -1.00 0.00 1.00 2.00 3.00 4.00
5 0.00 1.00 2.00 3.00 4.00 5.00
6

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209.00 1,163.00 39.00 17.00 7.00 Err:504
1 3,367.00 1,292.00 37.00 24.00 10.00 Err:504
2 3,871.00 1,474.00 53.00 22.00 5.77 Err:504
3 4,239.00 1,678.00 103.00 17.50 6.15 Err:504
4 4,929.00 1,865.00 56.01 19.36 6.56 Err:504
5 5,217.00 1,533.66 62.35 21.42 7.00 Err:504
6

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209.00 4,372.00 4,411.00 4,428.00 4,435.00 Err:504 Err:504
1 3,367.00 4,659.00 4,696.00 4,720.00 4,730.00 Err:504 Err:504
2 3,871.00 5,345.00 5,398.00 5,420.00 5,425.77 Err:504 Err:504
3 4,239.00 5,917.00 6,020.00 6,037.50 6,043.65 Err:504 Err:504
4 4,929.00 6,794.00 6,850.01 6,869.37 6,875.93 Err:504 Err:504
5 5,217.00 6,750.66 6,813.00 6,834.42 6,841.42 Err:504 Err:504
6

Loss Reserving Techniques in Non-Life 31 Chain ladder, Variant III

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Estimated Reserves
Year of origin Reserve
0 Err:504
1 Err:504
2 Err:504
3 Err:504
4 Err:504
5 Err:504
Total: Err:504

Loss Reserving Techniques in Non-Life 32 Chain ladder, Variant III

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method, Variant IV

The logarithm of the cumulative run-off triangle.
The cumulative run-off triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

ln(Cumulative triangle)
Development year
Year of origin 0 1 2 3 4 5 6
0 8.07 8.38 8.39 8.40 8.40 8.40
1 8.12 8.45 8.45 8.46 8.46
2 8.26 8.58 8.59 8.60
3 8.35 8.69 8.70
4 8.50 8.82
5 8.56
6

Estimation of the lineair least square parameters

Method 1 (using Excel functions)
column 0: a0 0.104684923
b0 8.050189709
column 1: a1 0.112066733
b1 8.360432289
column 2: a2 0.107227495
b2 8.374896106
column 3: a3 0.1010739
b3 8.383298922
column 4: a4 0.064397587
b4 8.397282895
column 5: a5 Err:504
b5 Err:504

Estimation of the cost level adjustment factors

j wj
w0 0 25 g0 1.110361
w1 1 16 g1 1.118588
w2 2 9 g2 1.113187
w3 3 4 g3 1.106358
w4 4 1 g4 1.066516
w5 5 0 g5 Err:504
sum 55 g Err:504

b0 0.10468492 g'0 Err:504 w0g0 27.75902

b1 0.11206673 g'1 Err:504 w1g1 17.8974
b2 0.10722749 g'2 Err:504 w2g2 10.01869

Loss Reserving Techniques in Non-Life 33 Chain ladder, Variant IV

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

b3 0.1010739 g'3 Err:504 w3g3 4.425434

b4 0.06439759 g'4 Err:504 w4g4 1.066516
b5 Err:504 g'5 Err:504 w5g5 Err:504
sum Err:504

Cost level adjustment of the incremental run-off triangle

The incremental run-off triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

n-(i+j) (n = 5)
Development year
Year of origin 0 1 2 3 4 5 6
0 5 4 3 2 1 0
1 4 3 2 1 0
2 3 2 1 0
3 2 1 0
4 1 0
5 0
6

x'-run-off triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
1 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504
2 Err:504 Err:504 Err:504 Err:504 Err:504
3 Err:504 Err:504 Err:504 Err:504
4 Err:504 Err:504 Err:504
5 Err:504 Err:504
6 Err:504

Completion of the x'-run-off triangle

Loss Reserving Techniques in Non-Life 34 Chain ladder, Variant IV

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

column 0
j 0
n-j 6
i wi,0 x'i,0 wi,0 x'i,0
0 1 Err:504 Err:504
1 2 Err:504 Err:504
2 3 Err:504 Err:504
3 4 Err:504 Err:504
4 5 Err:504 Err:504
5 6 Err:504 Err:504
6 7 Err:504 Err:504
28 Err:504
x'0 Err:504
column 1
j 1
n-j 5
i wi,01 x'i,1 wi,0 x'i,1
0 2 Err:504 Err:504
1 3 Err:504 Err:504
2 4 Err:504 Err:504
3 5 Err:504 Err:504
4 6 Err:504 Err:504
5 7 Err:504 Err:504
27 Err:504
x'1 Err:504
column 2
j 2
n-j 4
i wi,2 x'i,2 wi,2 x'i,2
0 3 Err:504 Err:504
1 4 Err:504 Err:504
2 5 Err:504 Err:504
3 6 Err:504 Err:504
4 7 Err:504 Err:504
25 Err:504
x'2 Err:504
column 3
j 3
n-j 3
i wi,3 x'i,3 wi,3 x'i,3
0 4 Err:504 Err:504
1 5 Err:504 Err:504
2 6 Err:504 Err:504
3 7 Err:504 Err:504
22 Err:504
x'3 Err:504
column 4

Loss Reserving Techniques in Non-Life 35 Chain ladder, Variant IV

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

j 4
n-j 2
i wi,4 x'i,4 wi,4 x'i,4
0 5 Err:504 Err:504
1 6 Err:504 Err:504
2 7 Err:504 Err:504
18 Err:504
x'4 Err:504
column 5
j 5
n-j 1
i wi,5 x'i,5 wi,5 x'i,5
0 6 Err:504 Err:504
1 7 Err:504 Err:504
13 Err:504
x'5 Err:504
column 6
j 6
n-j 0
i wi,6 x'i,6 wi,6 x'i,6
0 7 #VALUE! #VALUE!
7 #VALUE!
x'6 #VALUE!

The completed x'-run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
1 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
2 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
3 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
4 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
5 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
6 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!

Reversion of the cost-level adjustment

i+j-n
Development year
Year of origin 0 1 2 3 4 5 6
0 -6.00 -5.00 -4.00 -3.00 -2.00 -1.00 0.00
1 -5.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00
2 -4.00 -3.00 -2.00 -1.00 0.00 1.00 2.00
3 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00
4 -2.00 -1.00 0.00 1.00 2.00 3.00 4.00
5 -1.00 0.00 1.00 2.00 3.00 4.00 5.00
6 0.00 1.00 2.00 3.00 4.00 5.00 6.00

The completed incremental run-off triangle

Loss Reserving Techniques in Non-Life 36 Chain ladder, Variant IV

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Development year
Year of origin 0 1 2 3 4 5 6
0 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
1 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
2 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
3 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
4 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
5 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!
6 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504 #VALUE!

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 2,062.00 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504
1 2,031.00 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504
2 2,164.00 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504
3 2,320.00 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504
4 2,462.00 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504
5 2,651.00 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504
6 3,084.00 Err:504 Err:504 Err:504 Err:504 Err:504 Err:504

Estimated Reserves
Year of origin Reserve
0 0.0
1 Err:504
2 Err:504
3 Err:504
4 Err:504
5 Err:504
6 Err:504
Total: Err:504

Loss Reserving Techniques in Non-Life 37 Chain ladder, Variant IV

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method

Calculation of the proportionality factors.

The final loss run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 4,975 4,629 4,497 4,470 4,456 4,456
1 5,135 4,949 4,783 4,760 4,750
2 5,681 5,631 5,492 5,470
3 6,272 6,198 6,131
4 7,326 7,087
5 7,353
6

1.0314 1.0241 1.0049 1.0026 1.0000

pj 0.96955 0.97646 0.99513 0.99740 1.00000 DEVELOPMENT FAC

Completion of the cumulative run-off triangle

Method 1 (using the proportionality factors p i,i+1)
Development year
Year of origin 0 1 2 3 4 5 6
0 4,975 4,629.00 4,497.00 4,470.00 4,456.00 4,456.00 4,456.00
1 5,135 4,949.00 4,783.00 4,760.00 4,750.00 4,750.00 4,750.00
2 5,681 5,631.00 5,492.00 5,470.00 5,455.78 5,455.78 5,455.78
3 6,272 6,198.00 6,131.00 6,101.12 6,085.25 6,085.25 6,085.25
4 7,326 7,087.00 6,920.15 6,886.42 6,868.51 6,868.51 6,868.51
5 7,353 7,129.07 6,961.23 6,927.30 6,909.29 6,909.29 6,909.29
6

LATEST PALATEST IN LATEST P/IULT. PAID ULT. INC. ULT. P/I,

4,456.00 4,456.00 1 4,456.00 4,456.00 1
4,730.00 4,750.00 0.995789 4,752.40 4,750.00 1.0005046
5,420.00 5,470.00 0.990859 5,455.78 5,455.78 1.00000129
6,020.00 6,131.00 0.981895 6,086.06 6,085.25 1.00013342
6,794.00 7,087.00 0.958657 6,947.08 6,868.51 1.01143965
5,217 7,353 0.709506 7,366.66 6,909.29 1.06619617
TOTAL 32,637 35,247 0.925951 35,064 34,525 1.01561652

Loss Reserving Techniques in Non-Life 38 The Munich chain ladder method

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Calculation of the proportionality factors.

The final loss run-off triangle

Development year
Year of ori 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

0.7241 0.9887 0.9957 0.9981 0.9953

DEVELOPMENT FAC pj 1.38093 1.01143 1.00434 1.00186 1.00474 DEVELOPMENT FACTORS

Completion of the cumulative run-off triangle

Method 1 (using the proportionality factors p i,i+1)
Development year
Year of ori 0 1 2 3 4 5 6
0 3,209 4,372.00 4,411.00 4,428.00 4,435.00 4,456.00 4,456.00
1 3,367 4,659.00 4,696.00 4,720.00 4,730.00 4,752.40 4,752.40
2 3,871 5,345.00 5,398.00 5,420.00 5,430.07 5,455.78 5,455.78
3 4,239 5,917.00 6,020.00 6,046.15 6,057.38 6,086.06 6,086.06
4 4,929 6,794.00 6,871.67 6,901.52 6,914.34 6,947.08 6,947.08
5 5,217 7,204.33 7,286.69 7,318.34 7,331.94 7,366.66 7,366.66
6

Loss Reserving Techniques in Non-Life 39 The Munich chain ladder method

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

DEVELOPMENT FACTORS

Loss Reserving Techniques in Non-Life 40 The Munich chain ladder method

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The Vylder's Least Square Method.

Calculation of the proportionality factors.

The cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

The regression dataset 0 1 2 3

X line column L0 L1 L2 L3
3,209 0 0 1 0 0 0
3,367 1 0 0 1 0 0
3,871 2 0 0 0 1 0
4,239 3 0 0 0 0 1
4,929 4 0 0 0 0 0
5,217 5 0 0 0 0 0
1,163 0 1 1 0 0 0
1,292 1 1 0 1 0 0
1,474 2 1 0 0 1 0
1,678 3 1 0 0 0 1
1,865 4 1 0 0 0 0
39 0 2 1 0 0 0
37 1 2 0 1 0 0
53 2 2 0 0 1 0
103 3 2 0 0 0 1
17 0 3 1 0 0 0
24 1 3 0 1 0 0
22 2 3 0 0 1 0
7 0 4 1 0 0 0
10 1 4 0 1 0 0
21 0 5 1 0 0 0

RAPPORT DÉTAILLÉ

Statistiques de la régression

Loss Reserving Techniques in Non-Life 41 De Vylder regression (1)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Coefficient de 0.9955403258
Coefficient d 0.9911005403
Coefficient d 0.9822010806
Erreur-type 243.97432754
Observations 21

ANALYSE DE VARIANCE
Degré de liberté
Somme des carrés
Moyenne des carrés F Valeur critique de F
Régression 10 66289131.8 6628913.18 111.366372 6.8274985E-09
Résidus 10 595234.725 59523.4725
Total 20 66884366.6

Coefficients Erreur-type Statistique t Limite

Probabilité
inférieureLimite
pour supérieure
seuil de confiance
pour seuil
= 95%
de confiance = 95%
Constante 1685.825 377.525692 4.46545768 0.00120588 844.645341872 2527.00466
L0 -1664.825 288.100982 -5.7786162 0.00017807 -2306.7539874 -1022.89601
L1 -1605.825 288.100982 -5.5738269 0.00023613 -2247.7539874 -963.896013
L2 -1423.825 290.672087 -4.89838915 0.00062461 -2071.4827682 -776.167232
L3 -1169.825 296.305791 -3.94803286 0.002739 -1830.0354428 -509.614557
L4 -605.7 308.605826 -1.96269788 0.07808574 -1293.3166285 81.9166285
C0 3531.175 288.100982 12.2567267 2.39348E-07 2889.24601258 4173.10399
C1 1102.575 288.100982 3.82704354 0.00333494 460.646012577 1744.50399
C2 -161.75 290.672087 -0.55646898 0.59012478 -809.40776819 485.907768
C3 -100 296.305791 -0.33748918 0.74272519 -760.21044275 560.210443
C4 -42 308.605826 -0.13609594 0.89444621 -729.61662848 645.616628

ANALYSE DES RÉSIDUS RÉPARTITION DES PROBABILITÉS

Observation Prévisions X Résidus Centile X

1 3552.175 -343.175 2.38095238 7
2 3611.175 -244.175 7.14285714 10
3 3793.175 77.825 11.9047619 17
4 4047.175 191.825 16.6666667 21
5 4611.3 317.7 21.4285714 22
6 5217 0 26.1904762 24
7 1123.575 39.425 30.952381 37
8 1182.575 109.425 35.7142857 39
9 1364.575 109.425 40.4761905 53
10 1618.575 59.425 45.2380952 103
11 2182.7 -317.7 50 1163
12 -140.75 179.75 54.7619048 1292
13 -81.75 118.75 59.5238095 1474
14 100.25 -47.25 64.2857143 1678
15 354.25 -251.25 69.047619 1865
16 -79 96 73.8095238 3209
17 -20 44 78.5714286 3367
18 162 -140 83.3333333 3871
19 -21 28 88.0952381 4239
20 38 -28 92.8571429 4929
21 21 6.82121E-13 97.6190476 5217

The incured projected run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,552 1,124 (141) (79) (21) 21

Loss Reserving Techniques in Non-Life 42 De Vylder regression (1)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

1 3,611 1,183 (82) (20) 38 80

2 3,793 1,365 100 162 220 262
3 4,047 1,619 354 416 474 516
4 4,611 2,183 918 980 1,038 1,080
5 5,217 2,788 1,524 1,586 1,644 1,686
6

3,531 1102.575 -161.75 (100) (42) -

Estimated Reserves
Year of origin Reserve
0 0
1 80
2 482
3
4 4017
5 9228
Total:

Residuals
Development year
Year of origin 0 1 2 3 4 5 6
0 (343) 39 180 96 28 -
1 (244) 109 119 44 (28)
2 78 109 (47) (140)
3 192 59 (251)
4 318 (318)
5 -
6

Loss Reserving Techniques in Non-Life 43 De Vylder regression (1)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

4 0 1 2 3 4 5
L4 C0 C1 C2 C3 C4 C5
0 1 0 0 0 0 0
0 1 0 0 0 0 0
0 1 0 0 0 0 0
0 1 0 0 0 0 0
1 1 0 0 0 0 0
0 1 0 0 0 0 0
0 0 1 0 0 0 0
0 0 1 0 0 0 0
0 0 1 0 0 0 0
0 0 1 0 0 0 0
1 0 1 0 0 0 0
0 0 0 1 0 0 0
0 0 0 1 0 0 0
0 0 0 1 0 0 0
0 0 0 1 0 0 0
0 0 0 0 1 0 0
0 0 0 0 1 0 0
0 0 0 0 1 0 0
0 0 0 0 0 1 0
0 0 0 0 0 1 0
0 0 0 0 0 0 1

Loss Reserving Techniques in Non-Life 44 De Vylder regression (1)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

mite inférieure
Limite
poursupérieure
seuil de confiance
pour seuil= de
95,0%
confiance = 95,0%
844.645342 2527.00466
-2306.75399 -1022.89601
-2247.75399 -963.896013
-2071.48277 -776.167232
-1830.03544 -509.614557
-1293.31663 81.9166285
2889.24601 4173.10399
460.646013 1744.50399
-809.407768 485.907768
-760.210443 560.210443
-729.616628 645.616628

(1,665)

Loss Reserving Techniques in Non-Life 45 De Vylder regression (1)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

(1,606)
(1,424)
(1,170)
(606)
-

1,686

Loss Reserving Techniques in Non-Life 46 De Vylder regression (1)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The Vylder's Least Square Method.

Estimation of the parameters
The triangle of the model
Development year
Year of origin 0 1 2 3 4 5 6
0 p0S0 p1S0 p2S0 p3S0 p4S0 p5S0 p6S0
1 p0S1 p1S1 p2S1 p3S1 p4S1 p5S1
2 p0S2 p1S2 p2S2 p3S2 p4S2
3 p0S3 p1S3 p2S3 p3S3
4 p0S4 p1S4 p2S4
5 p0S5 p1S5
6 p0S6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Iteration p0 p1 p2 p3 p4 p5 S0 S1 S2 S3 S4 S5
1. 0.50000 0.30000 0.20000 0.10000 0.05000 0.04000 4984 5303 6090 6957 8894 10434
2. 0.56732 0.22930 0.01033 0.00384 0.00166 0.00421 5574 5892 6767 7451 8610 9196
3. 0.57067 0.21817 0.00934 0.00345 0.00149 0.00377 5586 5902 6780 7463 8626 9142
4. 0.57078 0.21778 0.00933 0.00344 0.00149 0.00376 5586 5903 6780 7463 8626 9140
5. 0.57078 0.21777 0.00933 0.00344 0.00149 0.00376 5586 5903 6780 7463 8626 9140
sum pj 0.80657
rescaled 0.70767 0.27000 0.01156 0.00427 0.00184 0.00466 4505 5468 6019 6956 6435 7372

The resulting parameters

p0r 0.70767 S0r 4,505.4751
p1r 0.27000 S1r 5,468.4782
p2r 0.01156 S2r 6,019.1588
p3r 0.00427 S3r 6,956.2030

Loss Reserving Techniques in Non-Life 47 De Vylder regression (2)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

p4r 0.00184 S4r 6,435.3532

p5r 0.00466 S5r 7,372.1232

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,188.37 1,216.46 52.10 19.24 8.30 21.00
1 3,869.86 1,476.47 63.24 23.35 10.07 25.49
2 4,259.55 1,625.15 69.61 25.70 11.09 28.06
3 4,922.67 1,878.15 80.44 29.70 12.81 32.42
4 4,554.08 1,737.52 74.42 27.48 11.85 30.00
5 5,217.00 1,990.45 85.25 31.48 13.58 34.36
6

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,188.37 4,404.83 4,456.94 4,476.18 4,484.48 4,505.48
1 3,869.86 5,346.33 5,409.57 5,432.92 5,442.99 5,468.48
2 4,259.55 5,884.71 5,954.31 5,980.02 5,991.10 6,019.16
3 4,922.67 6,800.82 6,881.26 6,910.97 6,923.78 6,956.20
4 4,554.08 6,291.60 6,366.02 6,393.50 6,405.36 6,435.35
5 5,217.00 7,207.45 7,292.70 7,324.18 7,337.76 7,372.12
6

Loss Reserving Techniques in Non-Life 48 De Vylder regression (2)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The linear Arithmetic Separation Method

In this example the seperation method starts from the non-cumulative triangle. We assume therefore that the
triangle that was given always has an estimate of one accident at the end of year 0.

Estimation of the parameters

The triangle of the model
Development year
Year of origin 0 1 2 3 4 5
0 r0l0 r1l1 r2l2 r3l3 r4l4 r5l5
1 r0l1 r1l2 r2l3 r3l4 r4l5
2 r0l2 r1l3 r2l4 r3l5
3 r0l3 r1l4 r2l5
4 r0l4 r1l5
5 r0l5
6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Calculation of r5 and l5
l5 7,238.00
r5 0.0029
Calculation of r4 and l4
l4 6,710.47
r4 0.0012
Calculation of r3 and l3
l3 5,790.86
r3 0.0032
Calculation of r2 and l2
l2 5,240.32
r2 0.0093
Calculation of r1 and l1
l1 4,606.46
r1 0.2526
Calculation of r0 and l0
l0 4,390.78
r0 0.7308

Loss Reserving Techniques in Non-Life 49 Linear Arithmetic Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Estimation of the parameters corresponding to the future calender years.

The values
resulting
The known
from the
values
lin.
l0 regression
0 4,391 4,156
l1 1 4,606 4,759
l2 2 5,240 5,361
l3 3 5,791 5,964
l4 4 6,710 6,567
l5 5 7,238 7,170
l6 6 7,773
l7 7 8,375
l8 8 8,978
l9 9 9,581
l 10 10 10,184
l 11 11 10,787
l 12 12 11,390

The regression parameters

a 602.82
b 4,155.77

The resulting parameters inflation

r0 0.7308 l0 4,391
r1 0.2526 l1 4,606 4.9%
r2 0.0093 l2 5,240 13.8%
r3 0.0032 l3 5,791 10.5%
r4 0.0012 l4 6,710 15.9%
r5 0.0029 l5 7,238 7.9%
l6 7,773 7.4%
l7 8,375 7.8%
l8 8,978 7.2%
l9 9,581 6.7%
l10 10,184 6.3%
l11 10,787 5.9%
l12 11,390 5.6%

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10 23
2 3,871 1,474 53 22 9 24
3 4,239 1,678 103 25 10 26
4 4,929 1,865 72 27 11 28
5 5,217 1,963 78 29 12 30
6

Loss Reserving Techniques in Non-Life 50 Linear Arithmetic Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730 4,753
2 3,871 5,345 5,398 5,420 5,429 5,454
3 4,239 5,917 6,020 6,045 6,055 6,081
4 4,929 6,794 6,866 6,893 6,904 6,932
5 5,217 7,180 7,258 7,286 7,298 7,328
6

Loss Reserving Techniques in Non-Life 51 Linear Arithmetic Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

od
sume therefore that the

Loss Reserving Techniques in Non-Life 52 Linear Arithmetic Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

73.08498
25.2551
0.928756
0.31916
0.121877
0.290135

Loss Reserving Techniques in Non-Life 53 Linear Arithmetic Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Loss Reserving Techniques in Non-Life 54 Linear Arithmetic Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The linear Arithmetic Separation Method

In this example the seperation method starts from the non-cumulative triangle. We assume therefore that
the triangle that was given always has an estimate of one accident at the end of year 0.

Estimation of the parameters

The triangle of the model
Development year
Year of origin 0 1 2 3 4 5 6
0 r0l0 r1l1 r2l2 r3l3 r4l4 r5l5
1 r0l1 r1l2 r2l3 r3l4 r4l5
2 r0l2 r1l3 r2l4 r3l5
3 r0l3 r1l4 r2l5
4 r0l4 r1l5
5 r0l5
6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Calculation of r5 and l5
l5 7,238.00
r5 0.0029
Calculation of r4 and l4
l4 6,710.47
r4 0.0012
Calculation of r3 and l3
l3 5,790.86
r3 0.0032
Calculation of r2 and l2
l2 5,240.32
r2 0.0093
Calculation of r1 and l1
l1 4,606.46
r1 0.2526
Calculation of r0 and l0
l0 4,390.78
r0 0.7308

Loss Reserving Techniques in Non-Life 55 Exp. Arithmetic separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Estimation of the parameters corresponding to the future calender years.

The values
The
resulting
known
from the lin.
values
regression
l0 0 4,391 4266.70513
l1 1 4,606 4746.21776
l2 2 5,240 5279.62029
l3 3 5,791 5872.96915
l4 4 6,710 6533.00137
l5 5 7,238 7267.21115
l6 6 8083.93492
l7 7 8992.446
l8 8 10003.0599
l9 9 11127.2513
l 10 10 12377.7848
l 11 11 13768.8592
l 12 12 15316.2691

The regression parameters

m 1.112385
b 4,266.71

The resulting parameters inflation

r0 0.7308 l0 4390.78
r1 0.2526 l1 4606.46 4.9%
r2 0.0093 l2 5240.32 13.8%
r3 0.0032 l3 5790.86 10.5%
r4 0.0012 l4 6710.47 15.9%
r5 0.0029 l5 7238.00 7.9%
l6 8083.93 11.7%
l7 8992.45 11.2%
l8 10003.06 11.2%
l9 11127.25 11.2%
l10 12377.78 11.2%
l11 13768.86 11.2%
l12 15316.27 11.2%

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10 23
2 3,871 1,474 53 22 10 26
3 4,239 1,678 103 26 11 29
4 4,929 1,865 75 29 12 32
5 5,217 2,042 84 32 14 36
6

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730 4,753
2 3,871 5,345 5,398 5,420 5,430 5,456
3 4,239 5,917 6,020 6,046 6,057 6,086
4 4,929 6,794 6,869 6,898 6,910 6,942

Loss Reserving Techniques in Non-Life 56 Exp. Arithmetic separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

5 5,217 7,259 7,342 7,374 7,388 7,424

6

Loss Reserving Techniques in Non-Life 57 Exp. Arithmetic separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The Linear Geometric Separation Method

In this example the seperation method starts from the non-cumulative triangle. We assume therefore that
the triangle that was given always has an estimate of one accident at the end of year 0.

Estimation of the parameters

The triangle of the model
Development year
Year of origin 0 1 2 3 4 5 6
0 r0l0 r1l1 r2l2 r3l3 r4l4 r5l5
1 r0l1 r1l2 r2l3 r3l4 r4l5
2 r0l2 r1l3 r2l4 r3l5
3 r0l3 r1l4 r2l5
4 r0l4 r1l5
5 r0l5
6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Calculation of r5 and l5
l5 129.10
r5 0.1627
Calculation of r4 and l4
l4 103.68
r4 0.0723
Calculation of r3 and l3
l3 82.46
r3 0.2011
Calculation of r2 and l2
l2 77.27
r2 0.5513
Calculation of r1 and l1
l1 71.46
r1 16.2619
Calculation of r0 and l0
l0 68.05
r0 47.1533

Loss Reserving Techniques in Non-Life 58 Linear Geometric Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Estimation of the parameters corresponding to the future calender years.

The values
The resulting
known from the
values lin.
l0 regression
0 68.1 59.6
l1 1 71.5 71.2
l2 2 77.3 82.9
l3 3 82.5 94.5
l4 4 103.7 106.1
l5 5 129.1 117.7
l6 6 129.4
l7 7 141.0
l8 8 152.6
l9 9 164.3
l 10 10 175.9
l 11 11 187.5
l 12 12 199.2

The regression parameters

a 11.63
b 59.59

The resulting parameters

r0 47.1533 l0 68.0547
r1 16.2619 l1 71.4611
r2 0.5513 l2 77.2735
r3 0.2011 l3 82.4585
r4 0.0723 l4 103.6776
r5 0.1627 l5 129.1011
l6 129.3777
l7 141.00822
l8 152.63869
l9 164.26916
l10 175.89964
l11 187.53011
l12 199.16058

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10 21
2 3,871 1,474 53 22 9 23
3 4,239 1,678 103 26 10 25
4 4,929 1,865 71 28 11 27
5 5,217 2,104 78 31 12 29
6

Loss Reserving Techniques in Non-Life 59 Linear Geometric Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730 4,751
2 3,871 5,345 5,398 5,420 5,429 5,452
3 4,239 5,917 6,020 6,046 6,056 6,081
4 4,929 6,794 6,865 6,894 6,905 6,931
5 5,217 7,321 7,399 7,429 7,441 7,470
6

Loss Reserving Techniques in Non-Life 60 Linear Geometric Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The Exponential Geometric Separation Method

In this example the seperation method starts from the non-cumulative triangle. We assume therefore that
the triangle that was given always has an estimate of one accident at the end of year 0.

Estimation of the parameters

The triangle of the model
Development year
Year of origin 0 1 2 3 4 5 6
0 r0l0 r1l1 r2l2 r3l3 r4l4 r5l5
1 r0l1 r1l2 r2l3 r3l4 r4l5
2 r0l2 r1l3 r2l4 r3l5
3 r0l3 r1l4 r2l5
4 r0l4 r1l5
5 r0l5
6

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Calculation of r5 and l5
l5 129.10
r5 0.1627
Calculation of r4 and l4
l4 103.68
r4 0.0723
Calculation of r3 and l3
l3 82.46
r3 0.2011
Calculation of r2 and l2
l2 77.27
r2 0.5513
Calculation of r1 and l1
l1 71.46
r1 16.2619
Calculation of r0 and l0
l0 68.05
r0 47.1533

Loss Reserving Techniques in Non-Life 61 Exp. Geometric Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Estimation of the parameters corresponding to the future calender years.

The values
The resulting
known from the
values lin.
l0 regression
0 68.1 63.146
l1 1 71.5 71.569
l2 2 77.3 81.117
l3 3 82.5 91.938
l4 4 103.7 104.202
l5 5 129.1 118.103
l6 6 133.857
l7 7 151.714
l8 8 171.953
l9 9 194.891
l 10 10 220.890
l 11 11 250.356
l 12 12 283.754

The regression parameters

m 1.13
b 63.15

The resulting parameters

r0 47.1533 l0 68.0547
r1 16.2619 l1 71.4611
r2 0.5513 l2 77.2735
r3 0.2011 l3 82.4585
r4 0.0723 l4 103.6776
r5 0.1627 l5 129.1011
l6 133.857
l7 151.714
l8 171.953
l9 194.891
l10 220.890
l11 250.356
l12 283.754

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10 22
2 3,871 1,474 53 22 10 25
3 4,239 1,678 103 27 11 28
4 4,929 1,865 74 31 12 32
5 5,217 2,177 84 35 14 36
6

Loss Reserving Techniques in Non-Life 62 Exp. Geometric Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730 4,752
2 3,871 5,345 5,398 5,420 5,430 5,454
3 4,239 5,917 6,020 6,047 6,058 6,086
4 4,929 6,794 6,868 6,898 6,911 6,942
5 5,217 7,394 7,477 7,512 7,526 7,562
6

Loss Reserving Techniques in Non-Life 63 Exp. Geometric Separation

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method (Mack estimation)

Calculation of the proportionality factors.

The cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

pj 1.38093 1.01143 1.00434 1.00186 1.00474

1.41205 1.02253 1.01097 1.00660 1.00474 1.00000

sums 24,832 27,087 20,525 14,568 9,165 4,456 -

19,615 27,087 20,525 14,568 9,165 4,456 -

The d-triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 1.362 1.009 1.004 1.002 1.005
1 1.384 1.008 1.005 1.002
2 1.381 1.010 1.004
3 1.396 1.017
4 1.378
5
6

Error term
1.100 0.028 0.001 0.000
0.026 0.057 0.003 0.000
0.000 0.012 0.000
0.943 0.211
0.032

0.5254 0.1026 0.0021 0.0007 0.0000

0.5254 0.1026 0.0021 0.0007 0.0002

Process Variance Multiplier

Annual 0.38905 0.10259 0.00211 0.00066 0.00021
Reserve 0.49462 0.10556 0.00298 0.00087 0.00021

Parameter Varia
Annual 0.00001 0.00000 0.00000 0.00000 0.00000
Reserve 0.00001 0.00000 0.00000 0.00000 0.00000

Loss Reserving Techniques in Non-Life 64 Mack coefficients

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Matrix of Estimation Error Factors

2 3 4 5 6
0.00000 0.00000 0.00000 0.00000 0.00001
2 0.00000 2.232096E-08 2.232E-08 2.232E-08 2.232E-08 2.2321E-08
3 0.00000 2.232096E-08 6.721E-08 6.721E-08 6.721E-08 6.72148E-08
4 0.00000 2.232096E-08 6.721E-08 1.678E-07 1.678E-07 1.67752E-07
5 0.00000 2.232096E-08 6.721E-08 1.678E-07 3.79E-06 0.00000379
6 0.00001 2.232096E-08 6.721E-08 1.678E-07 3.79E-06 1.37377E-05

Covariance Matrix - Estimation Error Only

2 3 4 5 6
4456 4752 5456 6086 6947
2 4456 0.4432034763 0.472684 0.542644 0.605333 0.69097208
3 4752 0.4726837526 1.518064 1.742748 1.944079 2.2191154
4 5456 0.5426441605 1.742748 4.99323 5.570074 6.35809338
5 6086 0.6053332619 1.944079 5.570074 140.3817 160.242046
6 6947 0.6909720811 2.219115 6.358093 160.242 663.007524

Correlation Matrix - Estimation Error Only

2 3 4 5 6
0.67 1.23 2.23 11.85 25.75
2 0.67 1.00 0.58 0.36 0.08 0.04
3 1.23 0.58 1.00 0.63 0.13 0.07
4 2.23 0.36 0.63 1.00 0.21 0.11
5 11.85 0.08 0.13 0.21 1.00 0.53
6 25.75 0.04 0.07 0.11 0.53 1.00

Loss Reserving Techniques in Non-Life 65 Mack coefficients

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Diagonal LDF Reserves Ultimate Proc Mult Proc Sdev Param Sdev
4,456 1.00000 0 4,456
4,730 1.00474 22 4,752 0.00021 0.99 4.42% 0.67
5,420 1.00660 36 5,456 0.00087 2.18 6.09% 1.23
6,020 1.01097 66 6,086 0.00298 4.26 6.44% 2.23
6,794 1.02253 153 6,947 0.10556 27.08 17.69% 11.85
5,217 1.41205 2,150 7,367 0.49462 60.36 2.81% 25.75

32,637 2,427

66.34 2.73% 34.22162

4456 4752.397 5455.784 6086.065 6947.084

4456 0.443203 0.472684 0.542644 0.605333 0.690972
4752.397 0.472684 1.518064 1.742748 1.944079 2.219115
5455.784 0.542644 1.742748 4.99323 5.570074 6.358093
6086.065 0.605333 1.944079 5.570074 140.3817 160.242
6947.084 0.690972 2.219115 6.358093 160.242 663.0075

Loss Reserving Techniques in Non-Life 66 Mack coefficients

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

Param Sdev Total Sdev

2.97% 1.19 5.33%

3.44% 2.50 6.99%
3.38% 4.81 7.28%
7.74% 29.56 19.31%
1.20% 65.63 3.05%

1.41% 75 3.08%

Loss Reserving Techniques in Non-Life 67 Mack coefficients

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

The chain ladder method (Mack estimation)

Calculation of the proportionality factors.

The cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730
2 3,871 5,345 5,398 5,420
3 4,239 5,917 6,020
4 4,929 6,794
5 5,217
6

The 1/d-triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 0.734 0.991 0.996 0.998 0.995
1 0.723 0.992 0.995 0.998
2 0.724 0.990 0.996
3 0.716 0.983
4 0.725
5
6

median 0.724 0.991 0.996 0.998 0.995

The 1/d-triangle
Development year
Year of origin 0 1 2 3 4 5 6
0 + + + + =
1 - + - -
2 = - =
3 - -
4 +
5
6

Diagonal N- N+ N n m E(N) Var(N)

0 - 1 - 1 0 - -
1 1 1 1 2 0 0.5000 0.75
2 - 2 - 2 0 0.5000 0.75
3 3 1 1 4 1 1.2500 3.44
4 2 1 1 3 1 0.7500 1.69
Total : 3 3.0000 6.63

Loss Reserving Techniques in Non-Life 68 Mack coefficients (2)

(E. Van den Borre)
 Fortis Bank Insurance
Corporate Actuarial Control

- - - - - 1 1 1 1 -
1 - 1 1 - 1 - -
- 1 - - - -
1 1 - -
- 1

inf sup
0 -
-1.19741 2.20
-1.19741 2.20
-2.383937 4.88
-1.796115 3.30
- 2.04 8.04

Loss Reserving Techniques in Non-Life 69 Mack coefficients (2)

(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

Christophides linear model

The incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10
2 3,871 1,474 53 22
3 4,239 1,678 103
4 4,929 1,865
5 5,217
6

Log (incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 8.07 7.06 3.66 2.83 1.95 3.04
1 8.12 7.16 3.61 3.18 2.30
2 8.26 7.30 3.97 3.09
3 8.35 7.43 4.63
4 8.50 7.53
5 8.56
6

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 70
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

0 1 2 3 4 5 1 2 3 4 5

Year of origin Development year Pij Yij a0 a1 a2 a3 a4 a5 b1 b2 b3 b4 b5

0 0 3,209 8.074 1 0 0 0 0 0 0 0 0 0 0
0 1 1,163 7.059 1 0 0 0 0 0 1 0 0 0 0
0 2 39 3.664 1 0 0 0 0 0 0 1 0 0 0
0 3 17 2.833 1 0 0 0 0 0 0 0 1 0 0
0 4 7 1.946 1 0 0 0 0 0 0 0 0 1 0
0 5 21 3.045 1 0 0 0 0 0 0 0 0 0 1
1 0 3,367 8.122 0 1 0 0 0 0 0 0 0 0 0
1 1 1,292 7.164 0 1 0 0 0 0 1 0 0 0 0
1 2 37 3.611 0 1 0 0 0 0 0 1 0 0 0
1 3 24 3.178 0 1 0 0 0 0 0 0 1 0 0
1 4 10 2.303 0 1 0 0 0 0 0 0 0 1 0
2 0 3,871 8.261 0 0 1 0 0 0 0 0 0 0 0
2 1 1,474 7.296 0 0 1 0 0 0 1 0 0 0 0
2 2 53 3.970 0 0 1 0 0 0 0 1 0 0 0
2 3 22 3.091 0 0 1 0 0 0 0 0 1 0 0
3 0 4,239 8.352 0 0 0 1 0 0 0 0 0 0 0
3 1 1,678 7.425 0 0 0 1 0 0 1 0 0 0 0
3 2 103 4.635 0 0 0 1 0 0 0 1 0 0 0
4 0 4,929 8.503 0 0 0 0 1 0 0 0 0 0 0
4 1 1,865 7.531 0 0 0 0 1 0 1 0 0 0 0
5 0 5,217 8.560 0 0 0 0 0 1 0 0 0 0 0

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 71
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

RAPPORT DÉTAILLÉ

Statistiques de la régression
Coefficient de d 0.9987258137
Coefficient de d 0.997453251
Coefficient de d 0.894906502
Erreur-type 0.1752879878
Observations 21

ANALYSE DE VARIANCE
Degré de libertéSomme des
Moyenne
carrés des carrésFValeur critique de F
Régression 11 120.3402 10.94002 391.657 1.1E-10
Résidus 10 0.307259 0.030726
Total 21 120.6475

Coefficients Erreur-typeStatistique
Limite inférieure
Probabilité
t Limite supérieure
pourLimite
seuil inférieure
de
pour
confiance
Limite
seuilpour
de
supérieure
=confiance
seuil
95% depour
confiance
= 95%
seuil de
= 95,0%
confiance = 95,0%
Constante 8.5006458854 0.135777 62.60718 2.6E-14 8.19811 8.803177 8.198115 8.80317699
a0 -0.5535203733 0.156237 -3.54282 0.00533 -0.90164 -0.2054 -0.901638 -0.2054026
a1 -0.393095772 0.156237 -2.51602 0.0306 -0.74121 -0.04498 -0.741214 -0.044978
a2 -0.2817166289 0.158676 -1.77542 0.10621 -0.63527 0.071836 -0.635269 0.07183607
a3 0.0368376361 0.163967 0.224665 0.82676 -0.3285 0.402179 -0.328503 0.40217865
a4 0 0 65535 #NUM! 0 0 0 0
a5 0.0590319176 0.221724 0.266241 0.79546 -0.435 0.553063 -0.434999 0.55306314
b1 -0.967384032 0.110862 -8.72603 5.5E-06 -1.2144 -0.72037 -1.2144 -0.7203684
b2 -4.2328969857 0.120809 -35.038 8.5E-12 -4.50208 -3.96372 -4.502076 -3.963718
b3 -5.0570984181 0.134197 -37.6842 4.1E-12 -5.35611 -4.75809 -5.356108 -4.7580893
b4 -5.9030901917 0.156237 -37.7829 4E-12 -6.25121 -5.55497 -6.251208 -5.5549724
b5 -4.9026030744 0.206992 -23.685 4.1E-10 -5.36381 -4.4414 -5.363809 -4.441397

ANALYSE DES RÉSIDUS

Observation Prévisions Yij Résidus

1 7.9471255121 0.126589
2 6.9797414801 0.079017

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 72
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

3 3.7142285264 -0.050667
4 2.890027094 -0.056814
5 2.0440353204 -0.098125
6 3.0445224377 -4.44E-16
7 8.1075501134 0.014227
8 7.1401660814 0.023781
9 3.8746531277 -0.263735
10 3.0504516953 0.127602
11 2.2044599217 0.098125
12 8.2189292565 0.042339
13 7.2515452245 0.04419
14 3.9860322708 -0.01574
15 3.1618308384 -0.070788
16 8.5374835215 -0.185401
17 7.5700994894 -0.144742
18 4.3045865357 0.330142
19 8.5006458854 0.002246
20 7.5332618534 -0.002246
21 8.559677803 0

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 73
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

Calculation of the variance-covariance matrix

The future design matrix

0 1 2 3 4 5 1 2 3 4 5
Year of origin i Development year j Pij a0 a1 a2 a3 a4 a5 b2 b3 b4 b5
1 5 ln(Ŷ1,5) 0 1 0 0 0 0 0 0 0 1
2 4 ln(Ŷ2,4) 0 0 1 0 0 0 0 0 1 0
2 5 ln(Ŷ2,5) 0 0 1 0 0 0 0 0 0 1
3 3 ln(Ŷ4,3) 0 0 0 1 0 0 0 1 0 0
3 4 ln(Ŷ4,4) 0 0 0 1 0 0 0 0 1 0
3 5 ln(Ŷ4,5) 0 0 0 1 0 0 0 0 0 1
4 3 ln(Ŷ4,3) 0 0 0 0 1 0 0 1 0 0
4 4 ln(Ŷ4,4) 0 0 0 0 1 0 0 0 1 0
4 5 ln(Ŷ4,5) 0 0 0 0 1 0 0 0 0 1
5 2 ln(Ŷ5,2) 0 0 0 0 0 1 1 0 0 0
5 3 ln(Ŷ5,3) 0 0 0 0 0 1 0 1 0 0
5 4 ln(Ŷ5,4) 0 0 0 0 0 1 0 0 1 0
5 5 ln(Ŷ5,5) 0 0 0 0 0 1 0 0 0 1

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 74
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

1 2 3 4 5 6 7 8 9 10
X= 1 1 1 0 0 0 0 0 0 0 1
2 1 0 1 0 0 0 0 0 1 0
3 1 0 1 0 0 0 0 0 0 1
4 1 0 0 1 0 0 0 1 0 0
5 1 0 0 1 0 0 0 0 1 0
6 1 0 0 1 0 0 0 0 0 1
7 1 0 0 0 1 0 0 1 0 0
8 1 0 0 0 1 0 0 0 1 0
9 1 0 0 0 1 0 0 0 0 1
10 1 0 0 0 0 1 1 0 0 0
11 1 0 0 0 0 1 0 1 0 0
12 1 0 0 0 0 1 0 0 1 0
13 1 0 0 0 0 1 0 0 0 1
14 1 0 0 0 0 0 0 0 0 0
15 1 0 0 0 0 0 0 0 0 0

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 75
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

X'= 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 1 0 0 0 0 0 0 0 0 0 0 0 0
3 0 1 1 0 0 0 0 0 0 0 0 0 0
4 0 0 0 1 1 1 0 0 0 0 0 0 0
5 0 0 0 0 0 0 1 1 1 0 0 0 0
6 0 0 0 0 0 0 0 0 0 1 1 1 1
7 0 0 0 0 0 0 0 0 0 1 0 0 0
8 0 0 0 1 0 0 1 0 0 0 1 0 0
9 0 1 0 0 1 0 0 1 0 0 0 1 0
10 1 0 1 0 0 1 0 0 1 0 0 0 1

1 2 3 4 5 6 7 8 9 10
(X'X)= 1 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
2 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
3 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
4 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
5 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
6 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
7 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
8 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
9 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
10 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###

1 2 3 4 5 6 7 8 9 10
(XT'X)-1 = 1 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
2 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
3 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
4 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
5 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
6 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
7 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
8 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
9 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
10 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 76
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

X (X'X)-1 = 1 2 3 4 5 6 7 8 9 10
1 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
2 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
3 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
4 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
5 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
6 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
7 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
8 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
9 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
10 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
11 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
12 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
13 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
14 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###
15 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ###

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 77
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
X (X'X)-1 X'= 1 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
2 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
3 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
4 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
5 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
6 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
7 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
8 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
9 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
10 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
11 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
12 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
13 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
14 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
15 #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###

s2 0.0307258787 (result of the regression analysis)

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 78
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

s 2 X (XT X)-1 X'= #VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###
#VALUE! #VALUE! ### ### #VALUE! #VALUE! #VALUE! ### ### ### ### ### ### ### ###

0 1 2 3 4 5 1 2 3 4 5
8.50064589 -1 0 0 0 0 0.1 -1 -4 -5 -6 -5

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 79
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

i j Yi,j var(Yi,j) E(xi,j) var(xi,j) s(xi,j) Constante a0 a1 a2 a3 a4 a5 b1 b2 b3 b4 b5

1 5 3.20 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 0 0 -5
2 4 2.32 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 0 -6 0
2 5 3.32 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 0 0 -5
3 3 3.48 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 -5 0 0
3 4 2.63 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 0 -6 0
3 5 3.63 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 0 0 -5
4 2 4.27 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 -4 0 0 0
4 3 3.44 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 -5 0 0
4 4 2.60 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 0 -6 0
4 5 3.60 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0 0 0 0 0 -5
5 1 7.59 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0.1 -1 0 0 0 0
5 2 4.33 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0.1 0 -4 0 0 0
5 3 3.50 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0.1 0 0 -5 0 0
5 4 2.66 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0.1 0 0 0 -6 0
5 5 3.66 #VALUE! ### ### #VALUE! 8.50064589 0 0 0 0 0 0.1 0 0 0 0 -5

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 80
(E. Van den Borre)
Fortis Bank Insurance
Coporate Actuarial Control

The completed incremental run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 1,163 39 17 7 21
1 3,367 1,292 37 24 10 #VALUE!
2 3,871 1,474 53 22 ### #VALUE!
3 4,239 1,678 103 ### ### #VALUE!
4 4,929 1,865 #VALUE! ### ### #VALUE!
5 5,217 #VALUE! #VALUE! ### ### #VALUE!
6

The completed cumulative run-off triangle

Development year
Year of origin 0 1 2 3 4 5 6
0 3,209 4,372 4,411 4,428 4,435 4,456
1 3,367 4,659 4,696 4,720 4,730 #VALUE!
2 3,871 5,345 5,398 5,420 ### #VALUE!
3 4,239 5,917 6,020 ### ### #VALUE!
4 4,929 6,794 #VALUE! ### ### #VALUE!
5 5,217 #VALUE! #VALUE! ### ### #VALUE!
6

Loss Reserving Techniques in Non-Life

Christophides (lin regr) 81
(E. Van den Borre)