UUM ETD | Universiti Utara Malaysian Electronic Theses and Dissertation
FAQs | Feedback | Search Tips | Sitemap

Toward an Optimal Loss Reserves Estimation Technique for Egyption General Insurance Business - An Empirical Study

Taha, Tarek Abd Elhamid Ahmed (2010) Toward an Optimal Loss Reserves Estimation Technique for Egyption General Insurance Business - An Empirical Study. PhD. thesis, Universiti Utara Malaysia.

[img] PDF
Tarek_Abd_Elhamid_Ahmed_Taha.pdf
Restricted to Registered users only

Download (3MB)

Abstract

The study is undertaken with dual aims. The first is to improve loss reserves estimation of the Egyptian general insurance market by applying and comparing the accuracy several actuarial techniques namely, modified Chain Ladder Technique (CLT) that takes the actual past and future inflation rates into consideration and use different types of mean to find the development factors, Taylor Separation Technique (TST) and Bornhuetter-Ferguson Technique (BFT) with some modifications. The second aim is to estimate reported loss reserve using cross sectional regression and time series regression techniques. The estimation techniques and forecasting models are evaluated based on an improved error calculation and an error evaluation, which have not being used before in the reserve estimation literature. Data for actuarial reserve estimation techniques is collected from the biggest governmental insurance company's (Misr Insurance Company) internal reports and consists of incremental paid claims within the last six years from 2003/2004 till 2008/2009. As for the forecasting models, the data are taken from both the company under investigation (27 yearly observations) and for the total market (151 yearly observations) for all tested variables. The tested variables include underwriting premiums, paid claim, reinsurance issued premiums,inflation rate, and the investment rate. The results of the study show that CLT that incorporates inflation rates adjustments and time value of money, and uses harmonic mean development factors is the best among all the techniques evaluated in this study. The cross sectional regression results show that only underwriting premium, paid claims and inflation rate are significantly related to the loss reserves, where the relationship is positive for underwriting premium and paid claims, and negative for inflation rate. For all the adopted time series models, the estimations are done by using the reported loss reserves data in fire insurance in the Egyptian market for the period 1981-2007. The series from 1981-2002 were used for the estimations process and the other remaining observations (data from 2003 till 2007) are used to evaluate the models as outside sample data. The models are evaluated based on their accuracy in predicting outside sample data. The study concludes that the best reported loss reserve estimation model in the case of fire insurance segment in Misr Insurance Company is the ARIMA model in all steps-ahead.

Item Type: Thesis (PhD.)
Additional Information: Call Number: HG8076 T128t 2010
Uncontrolled Keywords: Insurance, Egypt
Subjects: H Social Sciences > HG Finance
Divisions: College of Business (COB)
Depositing User: Mr Husni Ismail
Date Deposited: 14 Aug 2011 01:50
Last Modified: 24 Jul 2013 12:15
URI: http://etd.uum.edu.my/id/eprint/2410

Actions (login required)

View Item View Item