Estimating the Predictive Distribution for Loss Reserve Models Glenn Meyers ISO Innovative Analytics CAS Annual Meeting November 14, 2007.

Slides:



Advertisements
Similar presentations
Copula Representation of Joint Risk Driver Distribution
Advertisements

©Towers Perrin Emmanuel Bardis, FCAS, MAAA Cane Fall 2005 meeting Stochastic Reserving and Reserves Ranges Fall 2005 This document was designed for discussion.
Probability Distributions CSLU 2850.Lo1 Spring 2008 Cameron McInally Fordham University May contain work from the Creative Commons.
1 Regression Models & Loss Reserve Variability Prakash Narayan Ph.D., ACAS 2001 Casualty Loss Reserve Seminar.
Best Estimates for Reserves Glen Barnett and Ben Zehnwirth or find us on
An Introduction to Stochastic Reserve Analysis Gerald Kirschner, FCAS, MAAA Deloitte Consulting Casualty Loss Reserve Seminar September 2004.
Topic 2: Statistical Concepts and Market Returns
Bootstrap Estimation of the Predictive Distributions of Reserves Using Paid and Incurred Claims Huijuan Liu Cass Business School Lloyd’s of London 10/07/2007.
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 6-1 Chapter 6 The Normal Distribution and Other Continuous Distributions.
A Primer on the Exponential Family of Distributions David Clark & Charles Thayer American Re-Insurance GLM Call Paper
Math 116 Chapter 12.
CORRELATION & REGRESSION
Correlation Estimation for Property and Casualty Underwriting Losses Fred Klinker Insurance Services Office, Inc.
6.1 What is Statistics? Definition: Statistics – science of collecting, analyzing, and interpreting data in such a way that the conclusions can be objectively.
Integrating Reserve Risk Models into Economic Capital Models Stuart White, Corporate Actuary Casualty Loss Reserve Seminar, Washington D.C September.
RMTD 404 Lecture 8. 2 Power Recall what you learned about statistical errors in Chapter 4: Type I Error: Finding a difference when there is no true difference.
AP Statistics Chapter 9 Notes.
Bootstrapping Identify some of the forces behind the move to quantify reserve variability. Review current regulatory requirements regarding reserves and.
Prof. Dr. S. K. Bhattacharjee Department of Statistics University of Rajshahi.
A Beginner’s Guide to Bayesian Modelling Peter England, PhD EMB GIRO 2002.
2005 CLRS September 2005 Boston, Massachusetts
Two Approaches to Calculating Correlated Reserve Indications Across Multiple Lines of Business Gerald Kirschner Classic Solutions Casualty Loss Reserve.
More on Stochastic Reserving in General Insurance GIRO Convention, Killarney, October 2004 Peter England and Richard Verrall.
VI. Evaluate Model Fit Basic questions that modelers must address are: How well does the model fit the data? Do changes to a model, such as reparameterization,
The Examination of Residuals. Examination of Residuals The fitting of models to data is done using an iterative approach. The first step is to fit a simple.
The Common Shock Model for Correlations Between Lines of Insurance
1999 CASUALTY LOSS RESERVE SEMINAR Intermediate Track II - Techniques
Copyright © 2009 Cengage Learning 15.1 Chapter 16 Chi-Squared Tests.
1999 CAS SEMINAR ON RATEMAKING OPRYLAND HOTEL CONVENTION CENTER MARCH 11-12, 1999 MIS-43 APPLICATIONS OF THE MIXED EXPONENTIAL DISTRIBUTION CLIVE L. KEATINGE.
© 2007 Towers Perrin June 17, 2008 Loss Reserving: Performance Testing and the Control Cycle Casualty Actuarial Society Pierre Laurin.
Testing Models on Simulated Data Presented at the Casualty Loss Reserve Seminar September 19, 2008 Glenn Meyers, FCAS, PhD ISO Innovative Analytics.
Biostatistics, statistical software VII. Non-parametric tests: Wilcoxon’s signed rank test, Mann-Whitney U-test, Kruskal- Wallis test, Spearman’ rank correlation.
Estimating the Predictive Distribution for Loss Reserve Models Glenn Meyers Casualty Loss Reserve Seminar September 12, 2006.
2007 CAS Predictive Modeling Seminar Estimating Loss Costs at the Address Level Glenn Meyers ISO Innovative Analytics.
The Cost of Financing Insurance Version 2.0 Glenn Meyers Insurance Services Office Inc. CAS Ratemaking Seminar March 8, 2002.
Correlation Assume you have two measurements, x and y, on a set of objects, and would like to know if x and y are related. If they are directly related,
Reserve Variability – Session II: Who Is Doing What? Mark R. Shapland, FCAS, ASA, MAAA Casualty Actuarial Society Spring Meeting San Juan, Puerto Rico.
“The Effect of Changing Exposure Levels on Calendar Year Loss Trends” by Chris Styrsky, FCAS, MAAA Ratemaking Seminar March 10, 2005.
Statistical Decision Theory Bayes’ theorem: For discrete events For probability density functions.
Ranges of Reasonable Estimates Charles L. McClenahan, FCAS, MAAA Iowa Actuaries Club, February 9, 2004.
On Predictive Modeling for Claim Severity Paper in Spring 2005 CAS Forum Glenn Meyers ISO Innovative Analytics Predictive Modeling Seminar September 19,
CIA Annual Meeting LOOKING BACK…focused on the future.
Estimation and Application of Ranges of Reasonable Estimates Charles L. McClenahan, FCAS, MAAA 2003 Casualty Loss Reserve Seminar.
Stochastic Loss Reserving Using Bayesian MCMC Models Glenn Meyers Presented to Association des Actuaires I.A.R.D. November 6, 2015 Talk based on CAS Monograph.
Glenn Meyers ISO Innovative Analytics 2007 CAS Annual Meeting Estimating Loss Cost at the Address Level.
Correlation & Regression Analysis
The Cost of Financing Insurance with Emphasis on Reinsurance Glenn Meyers ISO CAS Ratemaking Seminar March 10, 2005.
One Madison Avenue New York Reducing Reserve Variance.
Portfolio wide Catastrophe Modelling Practical Issues.
Logistic regression. Recall the simple linear regression model: y =  0 +  1 x +  where we are trying to predict a continuous dependent variable y from.
Part 4 – Methods and Models. FileRef Guy Carpenter Methods and Models Method is an algorithm – a series of steps to follow  Chain ladder, BF, Cape Cod,
Stochastic Loss Reserving with the Collective Risk Model Glenn Meyers ISO Innovative Analytics Casualty Loss Reserving Seminar September 18, 2008.
2000 SEMINAR ON REINSURANCE PITFALLS IN FITTING LOSS DISTRIBUTIONS CLIVE L. KEATINGE.
A Stochastic Framework for Incremental Average Reserve Models Presented by Roger M. Hayne, PhD., FCAS, MAAA Casualty Loss Reserve Seminar September.
Session C7: Dynamic Risk Modeling Loss Simulation Model Working Party Basic Model Underlying Prototype Presented by Robert A. Bear Consulting Actuary and.
What’s the Point (Estimate)? Casualty Loss Reserve Seminar September 12-13, 2005 Roger M. Hayne, FCAS, MAAA.
Parameter Estimation. Statistics Probability specified inferred Steam engine pump “prediction” “estimation”
Basic Track II 2004 CLRS September 2004 Las Vegas, Nevada.
From “Reasonable Reserve Range” to “Carried Reserve” – What do you Book? 2007 CAS Annual Meeting Chicago, Illinois November 11-14, 2007 Mark R. Shapland,
September 11, 2001 Thomas L. Ghezzi, FCAS, MAAA Casualty Loss Reserve Seminar Call Paper Program Loss Reserving without Loss Development Patterns - Beyond.
1998 CASUALTY LOSS RESERVE SEMINAR Intermediate Track II - Techniques
Dave Clark American Re-Insurance 2003 Casualty Loss Reserve Seminar
Statistical Analysis Urmia University
Cost of Capital Issues April 16, 2002 John J. Kollar.
Materials for Lecture 18 Chapters 3 and 6
Managing Underwriting Risk & Capital
Basic Training for Statistical Process Control
Basic Training for Statistical Process Control
Where did we stop? The Bayes decision rule guarantees an optimal classification… … But it requires the knowledge of P(ci|x) (or p(x|ci) and P(ci)) We.
Presentation transcript:

Estimating the Predictive Distribution for Loss Reserve Models Glenn Meyers ISO Innovative Analytics CAS Annual Meeting November 14, 2007

S&P Report, November 2003 Insurance Actuaries – A Crisis in Credibility “Actuaries are signing off on reserves that turn out to be wildly inaccurate.”

Background to Methodology - 1 Zehnwirth/Mack –Loss reserve estimates via regression –y =  ∙x +  GLM – E[Y] = f(  ∙x) –Allows choice of f and the distribution of Y –Choices restricted to speed calculations Clark – Direct maximum likelihood –Assumes Y has an Overdispersed Poisson distribution

Background to Methodology - 2 Heckman/Meyers –Used Fourier transforms to calculate aggregate loss distributions in terms of frequency and severity distributions. Hayne –Applied Heckman/Meyers to calculate distributions of ultimate outcomes, given estimate of mean losses

High Level View of Paper Combine 1-2 above –Use aggregate loss distributions defined in terms of Fourier transforms to (1) estimate losses and (2) get distributions of ultimate outcomes. Uses “other information” from data of ISO and from other insurers. –Implemented with Bayes theorem

Objectives of Paper Develop a methodology for predicting the distribution of outcomes for a loss reserve model. The methodology will draw on the combined experience of other “similar” insurers. –Use Bayes’ Theorem to identify “similar” insurers. Illustrate the methodology on Schedule P data Test the predictions of the methodology on several insurers with data from later Schedule P reports. Compare results with reported reserves.

A Quick Description of the Methodology Expected loss is predicted by chain ladder/Cape Cod type formula The distribution of the actual loss around the expected loss is given by a collective risk (i.e. frequency/severity) model.

A Quick Description of the Methodology The first step in the methodology is to get the maximum likelihood estimates of the model parameters for several large insurers. For an insurer’s data –Find the likelihood (probability of the data) given the parameters of each model in the first step. –Use Bayes’ Theorem to find the posterior probability of each model in the first step given the insurer’s data.

A Quick Description of the Methodology The predictive loss model is a mixture of each of the models from the first step, weighted by its posterior probability. From the predictive loss model, one can calculate ranges or statistics of interest such as the standard deviation or various percentiles of the predicted outcomes.

The Data Commercial Auto Paid Losses from 1995 Schedule P (from AM Best) –Long enough tail to be interesting, yet we expect minimal development after 10 years. Selected 250 Insurance Groups –Exposure in all 10 years –Believable payment patterns –Set negative incremental losses equal to zero.

16 insurer groups account for one half of the premium volume

Look at Incremental Development Factors Accident year 1986 Proportion of loss paid in the “Lag” development year Divided the 250 Insurers into four industry segments, each accounting for about 1/4 of the total premium. Plot the payment paths

Incremental Development Factors Incremental development factors appear to be relatively stable for the 40 insurers that represent about 3/4 of the premium. They are highly unstable for the 210 insurers that represent about 1/4 of the premium. The variability appears to increase as size decreases

Do Incremental Development Factors Differ by Size of Insurer? Form loss triangles as the sum of the loss triangles for all insurers in each of the four industry segments defined above. Plot the payment paths

There is no consistent pattern in aggregate loss payment factors for the four industry segments. Segment 1 Segment 3 Segment 2 Segment 4

Expected Loss Model Paid Loss is the incremental paid loss in the AY and Lag ELR is the Expected Loss Ratio ELR and Dev Lag are unknown parameters –Can be estimated by maximum likelihood –Can be assigned posterior probabilities for Bayesian analysis Similar to “Cape Cod” method in that the expected loss ratio is estimated rather than determined externally.

Distribution of Actual Loss around the Expected Loss Compound Negative Binomial Distribution (CNB) –Conditional on Expected Loss – CNB(x | E[Paid Loss]) –Claim count is negative binomial –Claim severity distribution determined externally The claim severity distributions were derived from data reported to ISO. Policy Limit = $1,000,000 –Vary by settlement lag. Later lags are more severe. Claim Count has a negative binomial distribution with = E[Paid Loss]/E[Claim Severity] and c =.01 See Meyers “The Common Shock Model for Correlated Insurance Losses” for background on this model.

Claim Severity Distributions Lag  1 Lag 2 Lag 3 Lag 4 Lags 5-10

Where

Likelihood Function for a Given Insurer’s Losses – where

Maximum Likelihood Estimates Estimate ELR and Dev Lag simultaneously by maximum likelihood Constraints on Dev Lag –Dev 1 ≤ Dev 2 –Dev i ≥ Dev i+1 for i = 2,3,…,7 –Dev 8 = Dev 9 = Dev 10 Use R’s optim function to maximize likelihood –Read appendix of paper before you try this

Maximum Likelihood Estimates of Incremental Development Factors Loss development factors reflect the constraints on the MLE’s described in prior slide Contrast this with the observed 1986 loss development factors on the next slide

Incremental Development Factors (Repeat of Earlier Slide) Loss payment factors appear to be relatively stable for the 40 insurers that represent about 3/4 of the premium. They are highly unstable for the 210 insurers that represent about 1/4 of the premium. The variability appears to increase as size decreases

Maximum Likelihood Estimates of Expected Loss Ratios Estimates of the ELRs are more volatile for the smaller insurers.

Testing the Compound Negative Binomial (CNB) Assumption Calculate the percentiles of each observation given E[Paid Loss]. –55 observations for each insurer If CNB is right, the calculated percentiles should be uniformly distributed. Test with PP Plot –Sort calculated percentiles in increasing order –Vector (1:n)/(n+1) where n is the number of percentiles –The plot of the above two vectors against each other should be on the diagonal line.

Interpreting PP Plots Take 1000 lognormally distributed random variables with  = 0 and  = 2 as “data” If a whole bunch of predicted percentiles are at the ends, the predicted tail is too light. If a whole bunch of predicted percentiles are in the middle, the predicted tail is too heavy. If in general the predicted percentiles are low, the predicted mean is too high

Testing the CNB Assumptions Insurer Ranks 1-40 (Large Insurers) This sample has 55×40 or 2200 observations. According to the Kolomogorov-Smirnov test, D statistic for a sample of 2200 uniform random numbers should be within ± of the 45º line 95% of the time. Actual D statistic = As the plot shows, the predicted percentiles are slightly outside the 95% band. We are close.

Testing the CNB Assumptions Insurer Ranks 1-40 (Large Insurers) Breaking down the prior plot by settlement lag shows that there could be some improvement by settlement lag. But in general, not bad! pp plots by settlement lag

Testing the CNB Assumptions Insurer Ranks (Smaller Insurers) This is bad! pp plots by settlement lag

Using Bayes’ Theorem Let  = {ELR, Dev Lag, Lag = 1,2,…,10} be a set of models for the data. –A model may consist of different “models” or of different parameters for the same “model.” For each model in , calculate the likelihood of the data being analyzed.

Using Bayes’ Theorem Then using Bayes’ Theorem, calculate the posterior probability of each parameter set given the data.

Selecting Prior Probabilities For Lag, select the payment paths from the maximum likelihood estimates of the 40 largest insurers, each with equal probability. For ELR, first look at the distribution of maximum likelihood estimates of the ELR from the 40 largest insurers and visually “smooth out” the distribution. See the slide on ELR prior below. Note that Lag and ELR are assumed to be independent.

Prior Distribution of Loss Payment Paths Prior loss payment paths come from the loss development paths of the insurers ranked 1-40, with equal probability Posterior loss payment path is a mixture of prior loss development paths.

Prior Distribution of Expected Loss Ratios The prior distribution of expected loss ratios was chosen by visual inspection.

Predicting Future Loss Payments Using Bayes’ Theorem For each model, estimate the statistic of choice, S, for future loss payments. Examples of S –Expected value of future loss payments –Second moment of future loss payments –The probability density of a future loss payment of x, –The cumulative probability, or percentile, of a future loss payment of x. These examples can apply to single (AY,Lag) cells, of any combination of cells such as a given Lag or accident year.

Predicting Future Loss Payments Using Bayes’ Theorem for Sums over Sets of {AY,Lag} If we assume losses are independent by AY and Lag Actually use the negative multinomial distribution –Assumes correlation of frequency between lags in the same accident year

Predicting Future Loss Payments Using Bayes’ Theorem Calculate the Statistic S for each model. Then the posterior estimate of S is the model estimate of S weighted by the posterior probability of each model

Sample Calculations for Selected Insurers Coefficient of Variation of predictive distribution of unpaid losses. Plot the probability density of the predictive distribution of unpaid losses.

Predictive Distribution Insurer Rank 7 Predictive Mean = $401,951 K CV of Total Reserve = 6.9%

Predictive Distribution Insurer Rank 97 Predictive Mean = $40,277 K CV of Total Reserve = 12.6%

CV of Unpaid Losses

Validating the Model on Fresh Data Examined data from 2001 Annual Statements –Both 1995 and 2001 statements contained losses paid for accident years –Often statements did not agree in overlapping years because of changes in corporate structure. We got agreement in earned premium for 109 of the 250 insurers. Calculated the predicted percentiles for the amount paid Evaluate predictions with pp plots.

PP Plots on Validation Data KS 95% critical values = ±13.03%

Feedback If you have paid data, you must also have the posted reserves. How do your predictions match up with reported reserves? –In other words, is S&P right? Your results are conditional on the data reported in Schedule P. Shouldn’t an actuary with access to detailed company data (e.g. case reserves) be able to get more accurate estimates?

Response – Expand the Original Scope of the Paper Could persuade more people to look at the technical details. Warning – Do not over-generalize the results beyond commercial auto in timeframe.

Predictive and Reported Reserves For the validation sample, the predictive mean (in aggregate) is closer to the 2001 retrospective reserve. Possible conservatism in reserves. OK? “%” means % reported over the predictive mean. Retrospective = reported less paid prior to end of 1995.

Predictive Percentiles of Reported Reserves Conservatism is not evenly spread out. Conservatism appears to be independent of insurer size Except for the evidence of conservatism, the reserves are spread out in a way similar to losses. Were the reserves equal to ultimate losses?

Reported Reserves More Accurate? Divide the validation sample in to two groups and look at subsequent development. 1. Reported Reserve < Predictive Mean 2. Reported Reserve > Predictive Mean Expected result if Reported Reserve is accurate. –Reported Reserve = Retrospective Reserve for each group Expected result if Predictive Mean is accurate? –Predictive Mean  Retrospective Reserve for each group –There are still some outstanding losses in the retrospective reserve.

Subsequent Reserve Changes Group up/down Ups are bigger Group 2 More downs than ups Results are independent of insurer size Group 1Group 2

Subsequent Reserve Changes The CNB formula identified two groups where: –Group 1 tends to under-reserve –Group 2 tends to over-reserve Incomplete agreement at Group level –Some in each group get it right Discussion??

Main Points of Paper How do we evaluate stochastic loss reserve formula? –Test predictions of future loss payments –Test on several insurers –Main Focus Are there any formulas that can pass these tests? –Bayesian CNB does pretty good on CA Schedule P data. –Uses information from many insurers –Are there other formulas? This paper sets a bar for others to raise.

Subsequent Developments Paper completed in April 2006 Additional critique Describe recent developments Describe ongoing research

PP Plots on Validation Data Clive Keatinge’s Observation Does the leveling of plots at the end indicate that the predicted tails are too light? The plot is still within the KS bounds and thus is not statistically significant. The leveling looks rather systematic.

Alternative to the KS Anderson-Darling Test AD is more sensitive to tails. Critical values are 1.933, 2.492, and for 10, 5 and 1% levels respectively. Value for validation sample is Not outrageously bad, but Clive has a point. Explanation – Did not reflect all sources of uncertainty??

Is Bayesian Methodology Necessary? “Thinking Outside the Triangle” –Paper in June 2007 ASTIN Colloquium Works with simulated data on a similar model Compares Bayesian with maximum likelihood predictive distributions

Maximum Likelihood Fitting Methodology PP Plots for Combined Fits PP plot reveals the S-shape that characterizes overfitting. The tails are too light

Bayesian Fitting Methodology PP Plots for Combined Fits Nailed the Tails

IN THIS EXAMPLE Maximum Likelihood method understates the true variability I call this “overfitting” i.e. the model fits the data rather than the population Nine parameters fit to 55 points SPECULATION – Overfitting will occur in all maximum likelihood methods and in moment based methods –i.e. GLM and Mack

Expository Paper in Preparation Focus on the Bayesian method described in this paper Uses Gibbs sampler to simulate posterior distribution of the results Complete algorithm coded in R Hope to increase population of actuaries who: –Understand what the method means –Can actually use the method