Search In this Thesis
   Search In this Thesis  
العنوان
Modifying a probability model for claims data as a basis for actuarial operations in property insurance :
المؤلف
Saed, Abd El-Rahman El-Araby Mosad Mohammed.
هيئة الاعداد
باحث / عبدالرحمن العربي مسعد محمد سعيد
مشرف / محمد توفيق اسماعي
مشرف / جمال عبدالباقى واصف
مشرف / منى البشير الشربينى
مناقش / على السيد عبده
مناقش / عادل منير رابح
الموضوع
Data mining. Information storage and retrieval.
تاريخ النشر
2024.
عدد الصفحات
online resource (168 pages) :
اللغة
الإنجليزية
الدرجة
الدكتوراه
التخصص
الأعمال والإدارة والمحاسبة
تاريخ الإجازة
1/1/2024
مكان الإجازة
جامعة المنصورة - كلية التجارة - قسم الاحصاء التطبيقى والتأمين
الفهرس
Only 14 pages are availabe for public view

from 168

from 168

Abstract

Many methods and techniques are used to perform actuarial operations in property insurance companies; among these are probability distributions, which are used in making decisions in Premium Rating, Reservation, Reinsurance Agreements, and Testing for Solvency.This study aims to develop Probability distributions for Claims data for three branches of property insurance (Engineering, Marine Hull and Medical Insurance) in the Egyptian insurance market. This study introduces nine mixture probability distributions (Log .logistic Pareto, Log .logistic Lomax, Log .logistic jumble) for Engineering insurance. (Lindley Pareto, Lindely Lomax, Lindely Jumble) for Marine Hull insurance. (Weibull Pareto, Weibull Lomax, Weibull jumble) For Medical Insurance. The result of this study shows that (L. logistic jumble) distribution is the best distribution for Engineering insurance claims data, (Lindley Pareto distribution) is the best distribution for Marine Hull insurance claims data, and (Weibull Pareto) distribution is the best distribution for Medical Insurance claims data. The previous results were obtained after conducting several tests including Kolmogorov Smirnov (KS) statistics (with its p-value), Cramér-Von Mises (W), Anderson-Darling (A), the maximized log-likelihood (-ℓ̂), Akaike information criterion (AIC), Bayesian information criterion (BIC), Hannan-Quinn information criterion (HQIC), and consistent Akaike information criterion (CAIC). The R statistical package software is employed to calculate all of these measures.