Dissertation/Thesis Abstract

Estimation of severity distribution for hurricane losses in the United States: 1900-2005
by Yang, Fan, M.S., California State University, Long Beach, 2016, 70; 10004170
Abstract (Summary)

A typical problem that insurance and financial companies face is obtaining an appropriate probability distribution to model the magnitude (severity) of losses when they deal with forecasting future losses using past or current insurance claims data. Modeling the probability distribution of the magnitude of losses is often beneficial not only because it provides quantitative estimation of the expected loss, but also because it illustrates the risk associated with it, and can be used in making future inferences regarding premium rates, reserves, and expected profits. In the application section of this thesis, previous hurricane loss data set for the years 1900-2005 is used to find the best fitting probability distribution for the magnitude of the losses. The objective of this thesis is to find an appropriate probability distribution for the hurricane damages in the United States between 1900 and 2005, and to examine how well the selected probability distribution fits the hurricane loss data, applying the technique of Monte Carlo simulation. In addition, a non-parametric kernel estimation approach is performed to provide a smoothed fit of the empirical distribution for the hurricane loss data, and then the confidence intervals for the estimated characteristics are constructed, using bootstrap resampling method. To end with, the bootstrap confidence intervals for both the parametric and nonparametric cases are compared to the true values of the main statistical characteristics to access which model approach performs better in terms of the evaluation of the main statistics.

Indexing (document details)
Advisor: Moon, Hojin
Commitee: Kim, Sung Eun, Kim-Park, Yong Hee
School: California State University, Long Beach
Department: Mathematics and Statistics
School Location: United States -- California
Source: MAI 55/03M(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: American history, Statistics
Keywords: Hurricane losses, Loss distribution, United States
Publication Number: 10004170
ISBN: 9781339430645
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest