Citation

Kamaruzaman, Izzat Fakhruddin and Wan Zin, Wan Zawiah and Mohd Ariff, Noratiqah (2019) A generalized bivariate copula for flood analysis in Peninsular Malaysia. Malaysian Journal of Fundamental and Applied Sciences, 15 (1). pp. 38-49. ISSN 2289-5981, 2289-599X

Text 98.pdf - Published Version Restricted to Repository staff only Download (2MB)

Abstract

This study generalized the best copula to characterize the joint probability distribution between rainfall severity and duration in Peninsular Malaysia using two dimensional copulas. Specifically, to construct copulas, Inference Function for Margins (IFM) and Canonical Maximum Likelihood (CML) methods were specially exploited. For the purpose of achieving copula fitting, the derived rainfall variables by making use of the Standardized Precipitation Index (SPI) were fitted into several distributions. Five copulas, namely Gaussian, Clayton, Frank, Joe and Gumbel were put to the tests to establish the best data fitted copula. The tests produced acknowledged and satisfactory results of copula fitting for rainfall severity and duration. Surveying the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC), only three copulas produced a better fit for parametric and semi parametric approaches. Finally, two consistency tests were conducted and the results shown that Frank Copula produced consistent results.

Item Type:	Article
Uncontrolled Keywords:	Archimedean Copula, Elliptical Copula, Multivariate Distribution, Hydrology
Subjects:	P Language and Literature > P Philology. Linguistics
Divisions:	Faculty of Business (FOB)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	17 Feb 2022 01:22
Last Modified:	17 Feb 2022 01:22
URII:	http://shdl.mmu.edu.my/id/eprint/9154

Downloads

Edit (login required)