Generalized Classes of Distributions with Applications to Income and Lifetime Data

<p>In this thesis, new classes of distributions namely: exponentiated Kumaraswamy-Dagum (EKD), Log-exponentiated Kumaraswamy-Dagum (Log-EKD), McDonald Log-logistic (McLLog) and Gamma-Dagum (GD) distributions are presented. A thorough and comprehensive investigation of these classes of distributions is conducted. Mathematical properties of these classes of distributions including series expansion, hazard and reverse hazard functions, moments, generating functions, mean and median deviations, Bonferroni and Lorenz curves, distribution of order statistics, moments of order statistics and entropies are presented. Estimation of parameters of these distributions via maximum likelihood technique, Fisher information and asymptotic confidence intervals are given. Maximum likelihood estimation of the parameters of the exponentiated Kumaraswamy-Dagum distribution for censored data is constructed. Real data examples are presented to illustrate the usefulness and applicability of these proposed classes of distributions.</p>

Name thesis:degree_name

Master of Science in Mathematics (M.S.)

Level thesis:degree_level

Thesis (open access)

Discipline thesis:degree_discipline

Department of Mathematical Sciences

Year dc:date.available

2014

Author dc:creator

• Huang, Shujiao

Contributors dc:contributor

• Charles Champ
• Arpita Chatterjee

• ETD
• Kumaraswamy Distribution; Dagum Distribution; McDonald Distribution; Log-logistic Distribution; Gamma Distribution
• Applied Statistics
• Probability

Repository record dc:identifier

https://digitalcommons.georgiasouthern.edu/etd/1158

OAI identifier oai:identifier

oai:digitalcommons.georgiasouthern.edu:etd-2183