heteroskedasticity in loss reserving case fall 2012
DESCRIPTION
Heteroskedasticity in Loss Reserving CASE Fall 2012. What’s the proper link ratio?. And the boring version. All of those estimators are unbiased. Which one is efficient?. You’re not weighting link ratios. You’re making an assumption about the variance of the observed data. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/1.jpg)
Heteroskedasticity in Loss Reserving CASE Fall 2012
![Page 2: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/2.jpg)
What’s the proper link ratio?Cumulative
Paid @12 Months
Cumulative Paid @24
Months12-24
month LDF21,898 56,339 2.572822,549 59,459 2.636923,881 60,315 2.525625,897 71,409 2.757423,486 59,165 2.519227,029 60,778 2.248625,845 60,543 2.342525,415 54,791 2.155932,804 59,141 1.8029
![Page 3: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/3.jpg)
40,000 40,500 41,000 41,500 42,000 42,500 43,000 43,500 44,000 44,500 45,0000
10,000
20,000
30,000
40,000
50,000
60,000
Prior Cumulative Losses
Subs
eque
nt In
crem
enta
l
![Page 4: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/4.jpg)
0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,0000
10,000
20,000
30,000
40,000
50,000
60,000
Prior Cumulative Losses
Subs
eque
nt In
crem
enta
l
![Page 5: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/5.jpg)
0 5000 10000 15000 20000 25000 30000 35000 40000 450000
10000
20000
30000
40000
50000
60000
Prior Cumulative Losses
Incr
emen
tal
![Page 6: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/6.jpg)
Method LDFSimple average 2.3958Weighted average 2.3686Unweighted least squares 2.3375
And the boring version
All of those estimators are unbiased.Which one is efficient?
![Page 7: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/7.jpg)
You’re not weighting link ratios.
You’re making an assumption about the variance of the observed data.
![Page 8: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/8.jpg)
0 5000 10000 15000 20000 25000 30000 35000 40000 450000
10000
20000
30000
40000
50000
60000
Prior Cumulative Losses
Incr
emen
tal
![Page 9: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/9.jpg)
So how do we articulate our variance assumptions?
![Page 10: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/10.jpg)
A triangle is really a matrix
• A variable of interest (paid losses, for example) presumed to have some statistical relationship to one or more other variables in the matrix.
• The strength of that relationship may be established by creating models which relate two variables.
• A third variable is introduced by categorizing the predictors.• Development lag is generally used as the category.
![Page 11: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/11.jpg)
The response variable will generally be
incremental paid or incurred losses.
The design matrix may be either prior period
cumulative losses, earned premium or some other variable. Columns are
differentiated by category.
We assume that error terms
are homoskedastic and normally distributed
Calibrated model factors are
analogous to age-to-age
development factors.
ppnpn
p
n e
e
xx
xx
y
y
11
1
1111
![Page 12: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/12.jpg)
ppnpn
p
n e
e
xx
xx
y
y
11
1
1111
21,898 56,339 78,457 92,820 101,261 103,929 107,855 111,901 112,388 111,727 22,549 59,459 82,918 98,768 105,504 110,649 113,833 116,691 116,055 23,881 60,315 85,828 98,656 106,590 110,107 113,159 113,243 25,897 71,409 95,443 111,051 120,780 124,499 124,352 23,486 59,165 81,511 96,077 107,063 109,505 27,029 60,778 80,161 89,707 92,374 25,845 60,543 76,743 82,470 25,415 54,791 64,854 32,804 59,141 40,409
![Page 13: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/13.jpg)
Loss Reserving & Ordinary Least Squares Regression A Love Story
![Page 14: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/14.jpg)
Murphy variance assumptions
ebxy
exbxy
xebxy
LSM
WAD
SADMurphy: Unbiased Loss Development Factors
![Page 15: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/15.jpg)
Or, More Generally
exbxy 2/The multivariate model may be generally stated as containing a parameter to control the variance of the error term.
is not a hyperparameter. Fitting using SSE will always return = 0
![Page 16: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/16.jpg)
•Intuition•Losses vary in relation to predictors•Loss ratio variance looks different
•Observation•Behavior of a population•Diagnostics on individual sample
(Breusch-Pagan test)
![Page 17: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/17.jpg)
In 2011, Glenn Meyers & Peng Shi published NAIC Schedule P results for 132 companies. The object was to create a laboratory to determine which loss reserving method was most reliable.
![Page 18: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/18.jpg)
0 20 40 60 80
810
1214
16
Log MSE - PP Auto
Company
Ln(M
SE
)
810
1214
16
![Page 19: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/19.jpg)
0 20 40 60 80
05
1015
Log UpperMSE - PP Auto
Company
Ln(U
pper
MS
E)
0 5
1015
![Page 20: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/20.jpg)
0 20 40 60 80
-120
000
-800
00-4
0000
0
Upper Error - PP Auto
Company
Upp
er E
rror
-120
000
-800
00 -4
0000
0
![Page 21: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/21.jpg)
Observation of an Individual Sample
Breusch-Pagan Test•Use regression to diagnose your regression•Does the variance depend on the predictor?•Regress squared residuals against the predictor
![Page 22: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/22.jpg)
ii xe 102
An F-test determines the probability that the coefficients are non-zero.
![Page 23: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/23.jpg)
0.0 0.2 0.4 0.6 0.8 1.0
810
1214
16
BP vs MSE - PP Auto
BP pVal
Ln(M
SE
)
810
1214
16
![Page 24: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/24.jpg)
CaveatsCaveats•Non-normal error terms render B-P
meaningless!•Chain ladder utilizes stochastic
predictors•Earned premium has not been
adjusted
![Page 25: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/25.jpg)
Conclusions•Breusch-Pagan test is not strongly persuasive across the total data set.
•Homoskedastic error terms would support unweighted calibration of model factors.
•Probably more important to test functional form of error terms. Kolmogorov-Smirnov etc. may test for normal residuals.
![Page 26: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/26.jpg)
State your model and your underlying assumptions.
Test those assumptions.
Stop using models whose assumptions don’t reflect reality!
Statisticians have been doing this for years. Easy to steal leverage their work.
![Page 27: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/27.jpg)
-Dave Clark CAS Forum 2003
“Abandon your triangles!”
![Page 28: Heteroskedasticity in Loss Reserving CASE Fall 2012](https://reader036.vdocuments.site/reader036/viewer/2022062315/56815dc7550346895dcbf491/html5/thumbnails/28.jpg)
•https://
github.com/PirateGrunt/CASE-Spring-2013
•PirateGrunt.com
•http://lamages.blogspot.com/