PREMIUM PRICING AND RISK ASSESSMENT FOR CLAIM AMOUNTS BASED ON GENERALIZED LINEAR MODELS (GLM)
Öz
Actuarial Science is described as a mechanism that decreases the negative financial effects of random events which becomes obstacles to actualize reasonable expectations. It is important subject to make a fair share for the same amount of money which is paid by the people who has the same risk. It becomes even more important to be able to provide more effective methods with the reasonable prices on the customer retention and customer relationship management in the mutually competitive environment. In this case, it is expected to have methods which take into account customer’s previous claim experience with high predictive powers by insurance companies. Today, a large number of assumptions which may be used in the classical methods of analysis and predictions of this analysis are not sufficient. The main purpose of this study is of great importance for sustainable customer relationships, just make up a portfolio of premium pricing to be able to create a model that takes into account risk factors for individuals. GLM is a powerful methodology to evaluate the non-normal data. In this reason, it is formed an effective model that takes into account risk factors for the individuals in the portfolio using GLM. As a result of this analysis, it is chosen Logarithmic Gamma Model which gives the best results of the analysis for the customers that forms the data set. Finally, risk assessment was made by evaluating coefficient of variation, max, min and average of the claim amounts. At the end, 0.1% customers of the portfolio forms high risk group with regard to the change in the coefficient of variation.
Anahtar Kelimeler
Kaynakça
- Annette J. Dobson, A. G. (2008). An Introduction to Generalized Linear Models. Chapman&Hall, New York. Charles E. McCulloch, S. R. (2008). Generalized Linear Mixed and Mixed Models. Wiley, Canada.
- Edward W. Frees, V. R. (1999). A Longitudinal Data Analysis Interpretation of Credibility Models . Insurance: Mathematics and Economics, 229-247.
- Faraway, J. J. (2006). Extending the Linear Model with R:Generalized Linear, Mixed Effects and Nonpaametric Regression Models. Chapman&Hall, New York.
- Firth, D. (1988). Multiplicative Errors: Lognormal or Gamma? Journal of the Royal Statistical Society, 266-268.
- Frees, E. W. (2004). Longitudinal and Panel Data Analysis and Applications in the Social Sciences . Cambridge University Press, New York.
- Frees, E. W. (2010). Regression Modeling with Actuarial and Financial Applications . Cambridge University Press, New York.
- Gil, J. (2001). Generalized Linear Models: A Unified Approach. Sage Publications, New York.
- Hoffmann, J. P. (2004). Generalized Linear Models: An Applied Approach. Pearson Education, USA.
Ayrıntılar
Birincil Dil
Türkçe
Konular
-
Bölüm
Araştırma Makalesi
Yayımlanma Tarihi
1 Ekim 2014
Gönderilme Tarihi
10 Kasım 2013
Kabul Tarihi
-
Yayımlandığı Sayı
Yıl 2014