Risk Classification with Artificial Neural Networks Models in Motor Third Party Liability

Date

2019-06-29

Author

Yıldırak, Şahap Kasırga
Kestel, Sevtap Ayşe
Gür, İsmail

Metadata

Show full item record

Item Usage Stats

264
views

0
downloads

One of the most fundamental requirements in todays insurance sector is the determination of fair premium for the insured. In order for this purpose to be fulfilled, the correct risk classification is required for each insured in the portfolio. By the realization of correct risk classification, the insured can continue to be provided insurance services with more suitable pricing, while the insurance companies will have the opportunity to provide the correct person with the correct insurance and carry out financially sustainable insurance transactions. Risk classification as a common interest on both sides will ensure the existence and sustainability of the market. In this study, risk classification is made by using the claim information about insured within the scope of motor third party liability. As a classification model, ANN (Artificial Neural Network) models used on various attributes of the tools mentioned in the insurance policies are utilized. The data provided by the Insurance Information and Monitoring Center (SBM), consists of basic policy based information about insured vehicles, from 2006 to 2010. It has been shown that ANN models responds our problem significantly, model has reached high accuracy in classification process for both training and testing data.

URI

https://hdl.handle.net/11511/77536

Conference Name

International Conference on Data Science, Machine Learning and Statistics

Collections

Graduate School of Applied Mathematics, Conference / Seminar

Suggestions

OpenMETU
Core

Clustering Influence On MTPL Premium Estimation Using Credibility Approach Kızıloğlu, İsmail Onur; Kestel, Sevtap Ayşe; Yıldırım Külekci, Bükre; Department of Actuarial Sciences (2023-1-27) Pricing is one of the crucial aspects of the insurance industry. There are compulsory regulations on Motor Third Party Liability Insurance in Türkiye. After the cap premium, the significance of correct pricing for the profitability of the companies has increased even more. In this thesis, premiums are estimated quarterly using Bühlmann and Bühlmann-Straub credibility methods for the cities of Türkiye. First, the cities are sorted according to the total number of claims and divided into at least 3 and at mos...
Optimal premium allocation under stop-loss insurance using exposure curves Mert, Özenç Murat; Kestel, Sevtap Ayşe (2022-1-01) Determining the retention level in the stop-loss insurance risk premium for both insurer and reinsurer is an important factor in pricing. This paper aims to set optimal reinsurance with respect to the joint behavior of the insurer and the reinsurer under stop-loss contracts. The dependence between the costs of insurer and reinsurer is expressed as a function of retention (d) and maximum-cap (m) levels. Based on the maximum degree of correlation, the optimal levels for d and m are derived under certain claim...
Optimal premium allocation under stop-loss insurance using exposure curves Mert, Özenç Murat; Kestel, Sevtap Ayşe (2021-11-01) Determining the retention level in the stop-loss insurance risk premium for both insurer and reinsurer is an important factor in pricing. This paper aims to set optimal reinsurance with respect to the joint behavior of the insurer and the reinsurer under stop-loss contracts. The dependence between the costs of insurer and reinsurer is expressed as a function of retention (d) and maximum-cap (m) levels. Based on the maximum degree of correlation, the optimal levels for d and m are derived under certain claim...
Dependence Analysis with Normally Distributed Aggregate Claims in Stop Loss Insurance Mert, Özenç Murat; Kestel, Sevtap Ayşe (null; 2017-12-08) The reinsurance contracts in the insurance market have been playing an important role in the last couple of decades. One of the most important reinsurance contracts is the stop-loss reinsurance. It has an interesting property from the insurer point of view such that it is optimal if the criterion of minimizing the variance of the cost of the insurer is used. The word ‘optimality’ takes many researcher’s attention so that optimal reinsurance contracts under different assumptions have been investigated for de...
Impact of Outliers in Mortality Rates on the Valuation of Life Annuities Yavrum, Cem; Kestel, Sevtap Ayşe (2021-07-09) Annuity pricing is essential to insurance companies for their financial liabilities. Therefore, one of the purposes of companies is to adjust the annuity prices using a forecasting model that fits their historical data best. However, historical data may have outliers influencing the model. Extraordinary events such as a weak health system, an outbreak of war, and pandemics like Spanish flu or, more recently, Covid-19may cause outliers resulting in misevaluation of mortality rates. These outliers shoul...

Citation Formats

Ş. K. Yıldırak, S. A. Kestel, and İ. Gür, “Risk Classification with Artificial Neural Networks Models in Motor Third Party Liability,” presented at the International Conference on Data Science, Machine Learning and Statistics, Van, Türkiye, 2019, Accessed: 00, 2021. [Online]. Available: https://hdl.handle.net/11511/77536.