Thesis Type: Postgraduate
Institution Of The Thesis: Gazi University, Fen Bilimleri Enstitüsü, İSTATİSTİK ANABİLİM DALI, Turkey
Approval Date: 2015
Thesis Language: Turkish
Student: ESMA BİÇER
Supervisor: MERAL EBEGİL
Open Archive Collection: AVESIS Open Access Collection
Abstract:The purpose of credibility is finding results about past experience and the current experience. Since the beginning of 1900s, with the help of credibility theory, by using weighted average between these two experience, a fair premium (which) both insured and insurer comprimise has been tried. If insurance portfolio is homogeneous, to use general mean called collective premium representing collective experience of the insured is logical. However, insurance porfolios are generally heterogeneous. Therefore, insurance companies require to distinguish the insured having good or bad claims experience. That situation is also demanded by the insured. Whereat, to use only collective premium in the portfolio is not appropriate. Likely if insurance company determine to be taken premium values from the insured for the next term as an individual premium values, this condition will not be fair for the insurance company. For all these reasons, definding fair premium values between these two premiums are significant in terms of getting a deal between the insured and insurance company. Credibility theory is methods community, providing premium values to comprimise both the insured and insurance company via giving particular weights for these two points. One of the methods is Hachemeister Credibility Model. Initially, Hachemeister credibility model is introduced in this study. Then, to illustrate the applying of the Hachemeister credibility model, datum taken from Turkey Insurance Association site is used. Total amount of damage and injury and the premium value of provincial policy number (the number of the insured) datum are considered. Using amount of damage and the number of damage, amount of damage modeling have been studied with Hachemeister credibility model. Using premium value and the number of policy (the number of the insured), taken premium value modeling have been studied with Hachemeister credibility model. In the last part of the study, according to the number of different groups and periods, homogeneous and heterogeneous portfolios have been simulated. Hachmeister credibility model and premium value for the next insurance term have been obtained by using these portfolios. The resulting premium values, weighted collective premium values and weighted individual premium values are compared.Ultimately, results are interpreted.