Introducing the mixed distribution in fitting rainfall data

Several types of mixed distribution are proposed and tested in order to determine the best model in describing daily rainfall amount in Peninsular Malaysia for the time period of 33 years. A mixed distribution is a mixture of discrete and continuous daily rainfall which included the dry days. The mi...

Full description

Bibliographic Details
Main Authors: Syed Jamaludin, Sharifah Suhaila (Author), Yee, Kong Ching (Author), Yusof, Fadhilah (Author), Mean, Foo Hui (Author)
Format: Article
Language:English
Published: SciRes, 2011-10.
Subjects:
Online Access:Get fulltext
Description
Summary:Several types of mixed distribution are proposed and tested in order to determine the best model in describing daily rainfall amount in Peninsular Malaysia for the time period of 33 years. A mixed distribution is a mixture of discrete and continuous daily rainfall which included the dry days. The mixed distributions tested in this study were exponential distribution, gamma distribution, weibull distribution and lognormal distribution. The model will be selected based on the Akaike Information Criterion (AIC). In general, the mixed lognormal distribution has been selected as the best model for most of the rain gauge stations in Peninsular Malaysia. However, these results are greatly influenced by the topographical, geographical and climatic changes of the rain gauge stations.