Andrei L. Badescu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Andrei L. Badescu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Eric C. K. Cheung
School of Risk and Actuarial Studies, UNSW Business School, University of New South Wales, Sydney, NSW 2052, Australia
3 shared publications - X. Sheldon Lin
Department of Statistical Sciences, University of Toronto, 100 St George Street, Toronto, ONM5S 3G3, Canada
3 shared publications Alexandru V. Asimit
Department of Statistics , University of Toronto , Toronto, ON, Canada
1 shared publication- Dameng Tang
Department of Statistical Sciences University of Toronto100 St. George Street Toronto, ON M5S 3G3, Canada
1 shared publication David Landriault
University of Waterloo
1 shared publication- David Landriault
University of Waterloo
1 shared publication
Latest Publications
- 3 April 2020
- Published by Taylor & Francis in North American Actuarial Journal
- Vol. 25 (2) , 206-231
A New Class of Severity Regression Models with an Application to IBNR Prediction
Insurance loss severity data often exhibit heavy-tailed behavior, complex distributional characteristics such as multimodality, and peculiar links between policyholders’ risk profiles and claim amounts. To capture these features, we propose a transformed Gamma logit-weighted mixture of experts (TG-LRMoE) model for severity regression. The model possesses several desirable properties. The TG-LRMoE satisfies the denseness property that warrants its full versatility in capturing any distribution and regression structures. It may effectively extrapolate a wide range of tail behavior. The model is also identifiable, which further ensures its suitability for statistical inference. To make the TG-LRMoE computationally tractable, an expectation conditional maximization (ECM) algorithm with parameter penalization is developed for efficient and robust parameter estimation. The proposed model is applied to fit the severity and reporting delay components of a European automobile insurance dataset. In addition to obtaining excellent goodness of fit, the proposed model is shown to be useful and crucial for adequate prediction of incurred but not reported (IBNR) reserves through out-of-sample testing.
- 3 April 2020
- Published by Taylor & Francis in North American Actuarial Journal
- Vol. 25 (2) , 206-231
A New Class of Severity Regression Models with an Application to IBNR Prediction
Insurance loss severity data often exhibit heavy-tailed behavior, complex distributional characteristics such as multimodality, and peculiar links between policyholders’ risk profiles and claim amounts. To capture these features, we propose a transformed Gamma logit-weighted mixture of experts (TG-LRMoE) model for severity regression. The model possesses several desirable properties. The TG-LRMoE satisfies the denseness property that warrants its full versatility in capturing any distribution and regression structures. It may effectively extrapolate a wide range of tail behavior. The model is also identifiable, which further ensures its suitability for statistical inference. To make the TG-LRMoE computationally tractable, an expectation conditional maximization (ECM) algorithm with parameter penalization is developed for efficient and robust parameter estimation. The proposed model is applied to fit the severity and reporting delay components of a European automobile insurance dataset. In addition to obtaining excellent goodness of fit, the proposed model is shown to be useful and crucial for adequate prediction of incurred but not reported (IBNR) reserves through out-of-sample testing.
- 1 January 2019
- Published by Cambridge University Press (CUP) in ASTIN Bulletin
- Vol. 49 (03) , 647-688
A CLASS OF MIXTURE OF EXPERTS MODELS FOR GENERAL INSURANCE: APPLICATION TO CORRELATED CLAIM FREQUENCIES
This paper focuses on the estimation and application aspects of the Erlang count logit-weighted reduced mixture of experts model (EC-LRMoE), which is a fully flexible multivariate insurance claim frequency regression model. We first prove the identifiability property of the proposed model to ensure that it is a suitable candidate for statistical inference. An expectation conditional maximization (ECM) algorithm is developed for efficient model calibrations. Three simulation studies are performed to examine the effectiveness of the proposed ECM algorithm and the versatility of the proposed model. The applicability of the EC-LRMoE is shown through fitting an European automobile insurance data set. Since the data set contains several complex features, we find it necessary to adopt such a flexible model. Apart from showing excellent fitting results, we are able to interpret the fitted model in an insurance perspective and to visualize the relationship between policyholders’ information and their risk level. Finally, we demonstrate how the fitted model may be useful for insurance ratemaking.
- 1 January 2019
- Published by Cambridge University Press (CUP) in ASTIN Bulletin
- Vol. 49 (03) , 647-688
A CLASS OF MIXTURE OF EXPERTS MODELS FOR GENERAL INSURANCE: APPLICATION TO CORRELATED CLAIM FREQUENCIES
This paper focuses on the estimation and application aspects of the Erlang count logit-weighted reduced mixture of experts model (EC-LRMoE), which is a fully flexible multivariate insurance claim frequency regression model. We first prove the identifiability property of the proposed model to ensure that it is a suitable candidate for statistical inference. An expectation conditional maximization (ECM) algorithm is developed for efficient model calibrations. Three simulation studies are performed to examine the effectiveness of the proposed ECM algorithm and the versatility of the proposed model. The applicability of the EC-LRMoE is shown through fitting an European automobile insurance data set. Since the data set contains several complex features, we find it necessary to adopt such a flexible model. Apart from showing excellent fitting results, we are able to interpret the fitted model in an insurance perspective and to visualize the relationship between policyholders’ information and their risk level. Finally, we demonstrate how the fitted model may be useful for insurance ratemaking.
- 1 January 2019
- Published by Cambridge University Press (CUP) in ASTIN Bulletin
- Vol. 49 (03) , 709-739
A MARKED COX MODEL FOR THE NUMBER OF IBNR CLAIMS: ESTIMATION AND APPLICATION
Incurred but not reported (IBNR) loss reserving is of great importance for Property & Casualty (P&C) insurers. However, the temporal dependence exhibited in the claim arrival process is not reflected in many current loss reserving models, which might affect the accuracy of the IBNR reserve predictions. To overcome this shortcoming, we proposed a marked Cox process and showed its many desirable properties in Badescu et al. (2016).In this paper, we consider the model estimation and applications. We first present an expectation–maximization (EM) algorithm which guarantees the efficiency of the estimators unlike the moment estimation methods widely used in estimating Cox processes. In addition, the proposed fitting algorithm can be implemented at a reasonable computational cost. We examine the performance of the proposed algorithm through simulation studies. The applicability of the proposed model is tested by fitting it to a real insurance claim data set. Through out-of-sample tests, we find that the proposed model can provide realistic predictive distributions.
- 1 January 2019
- Published by Cambridge University Press (CUP) in ASTIN Bulletin
- Vol. 49 (03) , 709-739
A MARKED COX MODEL FOR THE NUMBER OF IBNR CLAIMS: ESTIMATION AND APPLICATION
Incurred but not reported (IBNR) loss reserving is of great importance for Property & Casualty (P&C) insurers. However, the temporal dependence exhibited in the claim arrival process is not reflected in many current loss reserving models, which might affect the accuracy of the IBNR reserve predictions. To overcome this shortcoming, we proposed a marked Cox process and showed its many desirable properties in Badescu et al. (2016).In this paper, we consider the model estimation and applications. We first present an expectation–maximization (EM) algorithm which guarantees the efficiency of the estimators unlike the moment estimation methods widely used in estimating Cox processes. In addition, the proposed fitting algorithm can be implemented at a reasonable computational cost. We examine the performance of the proposed algorithm through simulation studies. The applicability of the proposed model is tested by fitting it to a real insurance claim data set. Through out-of-sample tests, we find that the proposed model can provide realistic predictive distributions.
- 1 March 2018
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 79, 26-42
An IBNR–RBNS insurance risk model with marked Poisson arrivals
- 1 March 2018
- Published by Elsevier in Insurance: Mathematics and Economics
- Vol. 79, 26-42
An IBNR–RBNS insurance risk model with marked Poisson arrivals
- 1 May 2015
- Published by Cambridge University Press (CUP) in ASTIN Bulletin
- Vol. 45 (3) , 729-758
FITTING MIXTURES OF ERLANGS TO CENSORED AND TRUNCATED DATA USING THE EM ALGORITHM
We discuss how to fit mixtures of Erlangs to censored and truncated data by iteratively using the EM algorithm. Mixtures of Erlangs form a very versatile, yet analytically tractable, class of distributions making them suitable for loss modeling purposes. The effectiveness of the proposed algorithm is demonstrated on simulated data as well as real data sets.
- 1 May 2015
- Published by Cambridge University Press (CUP) in ASTIN Bulletin
- Vol. 45 (3) , 729-758
FITTING MIXTURES OF ERLANGS TO CENSORED AND TRUNCATED DATA USING THE EM ALGORITHM
We discuss how to fit mixtures of Erlangs to censored and truncated data by iteratively using the EM algorithm. Mixtures of Erlangs form a very versatile, yet analytically tractable, class of distributions making them suitable for loss modeling purposes. The effectiveness of the proposed algorithm is demonstrated on simulated data as well as real data sets.
- Open Access
- 1 September 2011
- Published by Cambridge University Press (CUP) in Journal of Applied Probability
- Vol. 48 (03) , 749-765
A Two-Dimensional Risk Model with Proportional Reinsurance
In this paper we consider an extension of the two-dimensional risk model introduced in Avram, Palmowski and Pistorius (2008a). To this end, we assume that there are two insurers. The first insurer is subject to claims arising from two independent compound Poisson processes. The second insurer, which can be viewed as a different line of business of the same insurer or as a reinsurer, covers a proportion of the claims arising from one of these two compound Poisson processes. We derive the Laplace transform of the time until ruin of at least one insurer when the claim sizes follow a general distribution. The surplus level of the first insurer when the second insurer is ruined first is discussed at the end in connection with some open problems.
- 1 September 2011
- Published by Cambridge University Press (CUP) in Journal of Applied Probability
- Vol. 48 (03) , 749-765
Open AccessA Two-Dimensional Risk Model with Proportional Reinsurance
In this paper we consider an extension of the two-dimensional risk model introduced in Avram, Palmowski and Pistorius (2008a). To this end, we assume that there are two insurers. The first insurer is subject to claims arising from two independent compound Poisson processes. The second insurer, which can be viewed as a different line of business of the same insurer or as a reinsurer, covers a proportion of the claims arising from one of these two compound Poisson processes. We derive the Laplace transform of the time until ruin of at least one insurer when the claim sizes follow a general distribution. The surplus level of the first insurer when the second insurer is ruined first is discussed at the end in connection with some open problems.
- Open Access
- 1 July 2011
- Published by Taylor & Francis in Stochastic Models
- Vol. 27 (3) , 407-430
On a Generalization of the Risk Model with Markovian Claim Arrivals
The class of risk models with Markovian arrival process (MAP) (see e.g., Neuts[ 15 Neuts , M.F. A versatile Markovian point process . Journal of Applied Probability 1979 , 16 ( 4 ), 764 – 779 . [Crossref], [Web of Science ®] [Google Scholar] ]) is generalized by allowing the waiting times between two successive events (which can be a change in the environmental state and/or a claim arrival) to have an arbitrary distribution. Using a probabilistic approach, we determine the solution for a class of Gerber–Shiu functions apart from some unknown constants when claim sizes have a mixed exponential distribution. Such constants are later determined using the more classic ruin-analytic approach. A numerical example is later considered to illustrate the tractability of the suggested methodology in the study of Gerber–Shiu functions.
- 1 July 2011
- Published by Taylor & Francis in Stochastic Models
- Vol. 27 (3) , 407-430
Open AccessOn a Generalization of the Risk Model with Markovian Claim Arrivals
The class of risk models with Markovian arrival process (MAP) (see e.g., Neuts[ 15 Neuts , M.F. A versatile Markovian point process . Journal of Applied Probability 1979 , 16 ( 4 ), 764 – 779 . [Crossref], [Web of Science ®] [Google Scholar] ]) is generalized by allowing the waiting times between two successive events (which can be a change in the environmental state and/or a claim arrival) to have an arbitrary distribution. Using a probabilistic approach, we determine the solution for a class of Gerber–Shiu functions apart from some unknown constants when claim sizes have a mixed exponential distribution. Such constants are later determined using the more classic ruin-analytic approach. A numerical example is later considered to illustrate the tractability of the suggested methodology in the study of Gerber–Shiu functions.
- 11 May 2010
- Published by Taylor & Francis in Scandinavian Actuarial Journal
- Vol. 2010 (2) , 93-104
Extremes on the discounted aggregate claims in a time dependent risk model
This paper presents an extension of the classical compound Poisson risk model for which the inter-claim time and the forthcoming claim amount are no longer independent random variables (rv's). Asymptotic tail probabilities for the discounted aggregate claims are presented when the force of interest is constant and the claim amounts are heavy tail distributed rv's. Furthermore, we derive asymptotic finite time ruin probabilities, as well as asymptotic approximations for some common risk measures associated with the discounted aggregate claims. A simulation study is performed in order to validate the results obtained in the free interest risk model.
- 11 May 2010
- Published by Taylor & Francis in Scandinavian Actuarial Journal
- Vol. 2010 (2) , 93-104
Extremes on the discounted aggregate claims in a time dependent risk model
This paper presents an extension of the classical compound Poisson risk model for which the inter-claim time and the forthcoming claim amount are no longer independent random variables (rv's). Asymptotic tail probabilities for the discounted aggregate claims are presented when the force of interest is constant and the claim amounts are heavy tail distributed rv's. Furthermore, we derive asymptotic finite time ruin probabilities, as well as asymptotic approximations for some common risk measures associated with the discounted aggregate claims. A simulation study is performed in order to validate the results obtained in the free interest risk model.
- 14 December 2007
- Published by Taylor & Francis in Scandinavian Actuarial Journal
- Vol. 2007 (4) , 227-247
Analysis of a threshold dividend strategy for a MAP risk model
We consider a class of Markovian risk models in which the insurer collects premiums at rate c1(c2) whenever the surplus level is below (above) a constant threshold level b. We derive the Laplace-Stieltjes transform (LST) of the distribution of the time to ruin as well as the LST (with respect to time) of the joint distribution of the time to ruin, the surplus prior to ruin, and the deficit at ruin. By interpreting that the insurer pays dividends continuously at rate c1−c2 whenever the surplus level is above b, we also derive the expected discounted value of total dividend payments made prior to ruin. Our results are obtained by making use of an existing connection which links an insurer's surplus process to an embedded fluid flow process.
- 14 December 2007
- Published by Taylor & Francis in Scandinavian Actuarial Journal
- Vol. 2007 (4) , 227-247
Analysis of a threshold dividend strategy for a MAP risk model
We consider a class of Markovian risk models in which the insurer collects premiums at rate c1(c2) whenever the surplus level is below (above) a constant threshold level b. We derive the Laplace-Stieltjes transform (LST) of the distribution of the time to ruin as well as the LST (with respect to time) of the joint distribution of the time to ruin, the surplus prior to ruin, and the deficit at ruin. By interpreting that the insurer pays dividends continuously at rate c1−c2 whenever the surplus level is above b, we also derive the expected discounted value of total dividend payments made prior to ruin. Our results are obtained by making use of an existing connection which links an insurer's surplus process to an embedded fluid flow process.