Bivariate hidden markov model to capture the dependency in claim estimate

Download

index.pdf

Date

2016

Author

Oflaz, Zarina

Metadata

Show full item record

Item Usage Stats

256
views

153
downloads

Most actuarial models rely on an assumption that both claim counts and aggregate claim amounts are serially independent, that simplifies the study of many risk quantities. However, this hypothesis does not always reflect the reality and is too restrictive in different frameworks. Some weather or economic conditions reasonably affect the claim-causing events, as a result, it influences both the claim number and the claim amount distributions. The unobservable background factor can be characterized by a hidden finite state Markov chain. In our study, we propose a novel approach for modeling claim dependence, Bivariate Hidden Markov Model (BHMM), which to our knowledge has not been studied before. We assume that the claim counts and the aggregate claim amounts are mutually dependent and serially dependent through an underlying hidden state. We construct three different Bivariate Hidden Markov Models, namely Poisson-Normal HMM, Poisson-Gamma HMM and Negative Binomial-Gamma HMM. To fit the model EM algorithm is used. In order to maximize the state-dependent part of complete-data log-likelihood of bivariate HMMs, we established and proved three propositions. In application part of our thesis, we fit the Poisson-Normal HMM with a different number of states to vehicle insurance observations for Istanbul taken from Traffic Insurances Information and Monitoring Center (TRAMER) for the years 2007-2009. In addition, we performed forecasting of distributions and state prediction, obtained the most likely sequence of states.

Subject Keywords

Actuarial science., Insurance, Insurance

URI

http://etd.lib.metu.edu.tr/upload/12620191/index.pdf
https://hdl.handle.net/11511/25827

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

Valuation of life insurance contracts using stochastic mortality rate and risk process modeling Çetinkaya, Şirzat; Hayfavi, Azize; Department of Financial Mathematics (2007) In life insurance contracts, actuaries generally value premiums using deterministic mortality rates and interest rates. They have ignored them stochastically in most of the studies. However it is known that neither interest rates nor mortality rates are constant. It is also known that companies may encounter insolvency problems such as ruin, so the ruin probability need to be added to the valuation of the life insurance contracts process. Insurance companies should model their surplus processes to price som...
Stochastic surplus processes with VaR AND CVaR simulations in actuarial applications Şimşek, Meral; Uğur, Ömür; Kestel, Sevtap Ayşe; Department of Actuarial Sciences (2016) The theory of ruin is a substantial study for those who are interested in ﬁnancial survival probability based on the patterns imposed by the surplus process, which determines the insurer’s capital balance at a given time. In other words, ﬂuctuations in aggregate claims as well as premiums in such processes can be secured by a certain capital. In this study, we simulate various surplus processes under different claim sizedistribution assumptions and extend the analyses by adding perturbation of a Brownian mo...
The Comparison of risk measures on claim distributions: Turkish motor insurance case Telkes, Cansu; Kestel, Sevtap Ayşe; Tank, Fatih; Department of Actuarial Sciences (2018) In this thesis, the impact of various risk measures on pricing methodology of automobile insurance product by using the historical claim data which is obtained from one of the most reputable insurance company in Turkey is investigated. To model the distribution of claim experience for pricing methodology, four right skewed distributions are chosen, namely Gamma, Weibull, Lognormal and Pareto. Two classical methods, which are methods of moment estimation and maximum likelihood estimation, are used to estimat...
Discrete-time surplus process for takaful insurance with multiple threshold levels Achlak, Arham; Kestel, Sevtap Ayşe; Tank, Fatih; Department of Actuarial Sciences (2016) Takaful is a type of insurance system whose participants contribute a certain sum of money to a common pool in order to guarantee each other against predefined losses. Over the past few years, the global Takaful industry has maintained solid growth trajectory with significant untapped potential across several markets. For this reason, there have been numerous researches conducted on the efficiency of Takaful system, but only few of them approach its actuarial aspect. This study aims to construct a new risk ...
Reinsurance pricing using exposure curve of two dependent risks Akarsu, Gülçin; Kestel, Sevtap Ayşe; Lourdes Centeno, Maria de; Department of Actuarial Sciences (2018) It is known that experience rating and exposure rating are used for insurance and reinsurance pricing by many practitioners. One of the main tools of exposure rating which is commonly used is exposure curves. It evaluates the percents of pure risk premium shared by insurer and reinsurer. In practice, the exposure curves which depend solely on claim history are widely utilized as an important indicator to determine the price and retention level based on the preferences and strategies of the insurer and reins...

Citation Formats

Z. Oflaz, “Bivariate hidden markov model to capture the dependency in claim estimate,” M.S. - Master of Science, Middle East Technical University, 2016.