Modern Actuarial Risk Theory Modern Actuarial Risk Theory by Rob Kaas University of Amsterdam, The Netherlands Marc Goovaerts Catholic University of Leuven, Belgium and University of Amsterdam, The Netherlands Jan Dhaene Catholic University of Leuven, Belgium and University of Amsterdam, The Netherlands and Michel Denuit Université Catholique de Louvain, Belgium KLUWER ACADEMIC PUBLISHERS NEW YORK, BOSTON, DORDRECHT, LONDON, MOSCOW eBookISBN: 0-306-47603-7 Print ISBN: 0-7923-7636-6 ©2002 Kluwer Academic Publishers NewYork, Boston, Dordrecht, London, Moscow Print ©2001 Kluwer Academic Publishers Dordrecht All rights reserved No part of this eBook maybe reproducedor transmitted inanyform or byanymeans,electronic, mechanical, recording, or otherwise, without written consent from the Publisher Created in the United States of America Visit Kluwer Online at: http://kluweronline.com and Kluwer's eBookstore at: http://ebooks.kluweronline.com Foreword Risk Theory has been identified and recognized as an important part of actuarial education; this is for example documented by the Syllabus of the Society of Actuaries and by the recommendations of the Groupe Consultatif. Hence it is desirable to have a diversity of textbooks in this area. I welcome the arrival of this new text in risk theory, which is original in several respects. In the language of figure skating or gymnastics, the text has two parts, the compulsory part and the free-style part. The compulsory part includes Chapters 1–4, which are compatible with official material of the Society of Actuaries. This feature makes the text also useful to students who prepare themselves for the actuarial exams. Other chapters are more of a free-style nature, for example Chapter 10 (Ordering of Risks, a speciality of the authors). And I would like to mention Chapter 8 in particular: to my knowledge, this is the first text in risk theory with an introduction to Generalized Linear Models. Special pedagogical efforts have been made throughout the book. The clear language and the numerous exercises are an example for this. Thus the book can be highly recommended as a textbook. I congratulate the authors to their text, and I would like to thank them also in the name of students and teachers that they undertook the effort to translate their text into English. I am sure that the text will be successfully used in many classrooms. H.U. Gerber Lausanne, October 3, 2001 v Preface This book gives a comprehensive survey of non-life insurance mathematics. It was originally written for use with the actuarial science programs at the Universities of Amsterdam and Leuven, but its Dutch version has been used at several other universities, as well as by the Dutch Actuarial Society. It provides a link to the further theoretical study of actuarial science. The methods presented can not only be used in non-life insurance, but are also effective in other branches of actuarial science, as well as, of course, in actuarial practice. Apart from the standard theory, this text contains methods that are directly rele- vant for actuarial practice, for instance the rating of automobile insurance policies, premium principles and IBNR models. Also, the important actuarial statistical tool of the Generalized Linear Models is presented. These models provide extra features beyond ordinary linear models and regression which are the statistical tools of choice for econometricians. Furthermore, a short introduction is given to credibility theory. Another topic which always has enjoyed the attention of risk theoreticians is the study of ordering of risks. The book reflects the state of the art in actuarial risk theory. Quite a lot of the results presented were published in the actuarial literature only in the last decade of the previous century. Models and paradigms studied An essential element of the models of life insurance is the time aspect. Between paying premiums and collecting the resulting pension, some decennia generally vii viii PREFACE elapse. This time-element is less prominently present in non-life insurance math- ematics. Here, however, the statistical models are generally more involved. The topics in the first five chapters of this textbook are basic for non-life actuarial science. The remaining chapters contain short introductions to some other topics traditionally regarded as non-life actuarial science. 1. The expected utility model The very existence of insurers can be explained by way of the expected utility model. In this model, an insured is a risk averse and rational decision maker, who by virtue of Jensen’s inequality is ready to pay more than the expected value of his claims just to be in a secure financial position. The mechanism through which decisions are taken under uncertainty is not by direct comparison of the expected payoffs of decisions, but rather of the expected utilities associated with these pay- offs. 2. The individual risk model In the individual risk model, as well as in the collective risk model that follows below, the total claims on a portfolio of insurance contracts is the random variable of interest. We want to compute, for instance, the probability that a certain capital will be sufficient to pay these claims, or the value-at-risk at level 95% associated with the portfolio, being the 95% quantile of its cumulative distribution function (cdf). The total claims is modelled as the sum of all claims on the policies, which are assumed independent. Such claims cannot always be modelled as purely dis- crete random variables, nor as purely continuous ones, and we provide a notation that encompasses both these as special cases. The individual model, though the most realistic possible, is not always very convenient, because the available data is used integrally and not in any way condensed. We study other techniques than convolution to obtain results in this model. Using transforms like the moment generating function helps in some special cases. Also, we present approximations based on fitting moments of the distribution. The Central Limit Theorem, which involves fitting two moments, is not sufficiently accurate in the important right- hand tail of the distribution. Hence, we also look at two more refined methods using three moments: the translated gamma approximation and the normal power approximation. 3. Collective risk models A model that is often used to approximate the individual model is the collective risk model. In this model, an insurance portfolio is viewed as a process that produces claims over time. The sizes of these claims are taken to be independent, identically distributed random variables, independent also of the number of claims generated. PREFACE ix This makes the total claims the sum of a random number of iid individual claim amounts. Usually one assumes additionally that the number of claims is a Poisson variate with the right mean. For the cdf of the individual claims, one takes an average of the cdf’s of the individual policies. This leads to a close fitting and computationally tractable model. Several techniques, including Panjer’s recursion formula, to compute the cdf of the total claims modelled this way are presented. 4. The ruin model In the ruin model the stability of an insurer is studied. Starting from capital at time his capital is assumed to increase linearly in time by fixed annual premiums, but it decreases with a jump whenever a claim occurs. Ruin occurs when the capital is negative at some point in time. The probability that this ever happens, under the assumption that the annual premium as well as the claim gen- erating process remain unchanged, is a good indication of whether the insurer’s assets are matched to his liabilities sufficiently well. If not, one may take out more reinsurance, raise the premiums or increase the initial capital. Analytical methods to compute ruin probabilities exist only for claims distribu- tions that are mixtures and combinations of exponential distributions. Algorithms exist for discrete distributions with not too many mass points. Also, tight upper and lower bounds can be derived. Instead of looking at the ruin probability, often one just considers an upper bound for it with a simple exponential structure (Lundberg). 5. Premium principles Assuming that the cdf of a risk is known, or at least some characteristics of it like mean and variance, a premium principle assigns to the risk a real number used as a financial compensation for the one who takes over this risk. Note that we study only risk premiums, disregarding surcharges for costs incurred by the insurance company. By the law of large numbers, to avoid eventual ruin the total premium should be at least equal to the expected total claims, but additionally, there has to be a loading in the premium to compensate the insurer for being in a less safeposi- tion. From this loading, the insurer has to build a reservoir to draw upon in adverse times, so as to avoid getting in ruin. We present a number of premium principles, together with the most important properties that can be attributed to premium principles. The choice of a premium principle depends heavily on the importance attached to such properties. There is no premium principle which is uniformly best. 6. Bonus-malus systems With some types of insurance, notably car insurance, charging a premium based exclusively on factors known a priori is insufficient. To incorporate the effect of risk factors of which the use as rating factors is inappropriate, such as race or quite x PREFACE often sex of the policy holder, and also of non-observable factors, such as state of health, reflexes and accident proneness, many countries apply an experience rating system. Such systems on the one hand use premiums based on a priori factors such as type of coverage and catalogue price or weight of a car, on the other hand they adjust these premiums by use of some kind of bonus-malus system, where one gets more discount after a claim-free year, but pays a higher premium after filing one or more claims. In this way, premiums are charged that reflect the exact driving capabilities of the driver better. The situation can be modelled as a Markov chain. 7. Credibility theory The claims experience on a policy may vary by two different causes. The first is the quality of the risk, expressed through a risk parameter. This represents the average annual claims in the hypothetical situation that the policy is monitored without change over a very long period of time. The other is the purely random good and bad luck of the policyholder that results in yearly deviations from the risk parameter. Credibility theory assumes that the risk quality is a drawing from a certain structure distribution, and that conditionally given the risk quality, the actual claims experience is a sample from a distribution having the risk quality as its mean value. The predictor for next year’s experience that is linear in the claims experience and optimal in the sense of least squares turns out to be a weighted average of the claims experience of the individual contract and the experience for the whole portfolio. The weight factor is the credibility attached to the individual experience, hence it is called the credibility factor, and the resulting premiums are called credibility premiums. As a special case, we study a bonus-malus system for car insurance based on a gamma-Poisson mixture model. 8. Generalized linear models Many problems in actuarial statistics can be written as Generalized Linear Mod- els (GLM). Instead of assuming the error term to be normally distributed, other types of randomness are allowed as well, such as Poisson, gamma and binomial. Moreover, the expected value of the dependent variable is not necessarily linear in the regressors, but it may also be equal to a function of a linear form of the covariates, for instance the logarithm. In this last case, one gets the multiplicative models which are appropriate in most insurance situations. This way, one can for instance tackle the problem of estimating the reserve to be kept for IBNR claims, see below. But one can also easily estimate the premiums to be charged for drivers from region in bonus class with car weight In credibility models, there are random group effects, but in GLM’s the effects are fixed, though unknown. For the latter class of problems, software is available that can handle a multitude of models. PREFACE xi 9. IBNR techniques An important statistical problem for the practicing actuary is the forecasting of the total of the claims that are Incurred, But Not Reported, hence the acronym IBNR, or not fully settled. Most techniques to determine estimates for this total are based on so-called run-off triangles, in which claim totals are grouped by year of origin and development year. Many traditional actuarial reserving methods turn out to be maximum likelihood estimations in special cases of GLM’s. 10. Ordering of risks It is the very essence of the actuary’s profession to be able to express preferences between random future gains or losses. Therefore, stochastic ordering is a vital part of his education and of his toolbox. Sometimes it happens that for two losses X and Y, it is known that every sensible decision maker prefers losing X, because Y is in a sense ‘larger’ than X. It may also happen that only the smaller group of all risk averse decision makers agree about which risk to prefer. In this case, risk Y may be larger than X, or merely more ‘spread’, which also makes a risk less at- tractive. When we interpret ‘more spread’ as having thicker tails of the cumulative distribution function, we get a method of ordering risks that has many appealing properties. For instance, the preferred loss also outdoes the other one as regards zero utility premiums, ruin probabilities, and stop-loss premiums for compound distributions with these risks as individual terms. It can be shown that the collec- tive model of Chapter 3 is more spread than the individual model it approximates, hence using the collective model, as a rule, leads to more conservative decisions regarding premiums to be asked, reserves to be held, and values-at-risk. Also, we can prove that the stop-loss insurance, proven optimal as regards the variance of the retained risk in Chapter 1, is also preferable, other things being equal, in the eyes of all risk averse decision makers. Sometimes, stop-loss premiums have to be set under incomplete information. We give a method to compute the maximal possible stop-loss premium assuming that the mean, the variance and an upper bound for a risk are known. In the individual and the collective model, as well as in ruin models, we assume that the claim sizes are stochastically independent non-negative random variables. Sometimes this assumption is not fulfilled, for instance there is an obvious depen- dence between the mortality risks of a married couple, between the earthquake risks of neighboring houses, and between consecutive payments resulting from a life insurance policy, not only if the payments stop or start in case of death, but also in case of a random force of interest. We give a short introduction to the risk ordering that applies for this case. It turns out that stop-loss premiums for a sum of random variables with an unknown joint distribution but fixed marginals are xii PREFACE maximal if these variables are as dependent as the marginal distributions allow, making it impossible that the outcome of one is ‘hedged’ by another. Educational aspects As this text has been in use for more than a decade at the University of Amsterdam and elsewhere, we could draw upon a long series of exams, resulting in long lists of exercises. Also, many examples are given, making this book well-suited as a textbook. Some less elementary exercises have been marked by and these might be skipped. The required mathematical background is on a level such as acquired in the first stage of a bachelors program in quantitative economics (econometrics or actuarial science), or mathematical statistics, making it possible to use the book either in the final year of such a bachelors program, or in a subsequent masters program in either actuarial science proper or in quantitative financial economics with a strong insurance component. To make the book accessible to non-actuaries, notation and jargon from life insurance mathematics is avoided. Therefore also students in applied mathematics or statistics with an interest in the stochastic aspects of insurance will be able to study from this book. To give an idea of the mathematical rigor and statistical sophistication at which we aimed, let us remark that moment generating functions are used routinely, while characteristic functions and measure theory are avoided. Prior experience with regression models is not required, but helpful. As a service to the student help is offered, in a separate section at the end of the book, with most of the exercises. It takes the form of either a final answer to check one’s work, or a useful hint. There is an extensive index, and the tables that might be needed on an exam are printed in the back. The list of references is not a thorough justification with bibliographical data on every result used, but more a list of useful books and papers containing more details on the topics studied, and suggesting further reading. Ample attention is given to computing techniques, but there is also attention for old fashioned approximation methods like the Central Limit Theorem (CLT). These methods are not only fast, but also often prove to be surprisingly accurate, and moreover they provide solutions of a parametric nature such that one does not have to recalculate everything after a minor change in the data. Also, we want to stress that ‘exact’ methods are as exact as their input. The order of magnitude of errors resulting from inaccurate input is often much greater than the one caused by using an approximate method. The notation used in this book conforms to what is usual in mathematical statistics as well as non-life insurance mathematics. See for instance the book by Bowers et al. (1986), the non-life part of which is similar in design to the first part of this book.