An early warning model for Turkish insurance companies

---

Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Uygulamalı Matematik Enstitüsü, Aktüerya Bilimleri Anabilim Dalı, Türkiye

Tezin Onay Tarihi: 2015

Öğrenci: GİZEM OCAK

Danışman: SEVTAP AYŞE KESTEL

Özet:

In Turkey, insurance companies have some obligations to be solvent and for this reason they are regularly audited according to defined constraints. There are some models and methods in order to analyze company performances. One of them is an early warning model that is constructed by using some financial ratios. The aim of the study is to determine how Solvency requirements affected the financial stability of Turkish insurance companies last years. Firstly, the proposed model takes into account the financial ratios related to liquidity, profitability, and other factors regarding to the country specific properties which was also used in study done by Genc (2004). Historical data on companies’ financial indicators are evaluated based on comparative linear model estimation methods to determine the company’s financial position which functions as an early warning indicator. In this study, used four methods have been employed to construct the predictor model as an early warning system which are linear regression, Multiperod Discriminant analysis, Logistic and Bayesian Regression. Financial details of 41 insurance companies which acted in the period of 1998-2012 in Turkish market was used. After determination of best fitted model, 2013 prediction was applied to all existence insurance companies.