Pemodelan Taburan Kebarangkalian Zarah Terampai Melampau di Lembah Klang

Abstract

This study aims to identify the best statistical model to represent the data set for one of the air pollutants that is the particulate matter with diameters smaller than 10 micrometers (PM10). Data from six air quality monitoring stations in the Klang Valley from 2009 to 2011 were used in this study. In determining the more appropriate probability distribution,both parametric and non-parametric approaches were tested. Two series of extreme data for PM10 were used, which are the monthly maximum and the Peak over threshold data series. Next, two parametric distributions, which are the Generalized Extreme Value (GEV) and Generalized Pareto (GPD) were fitted to the monthly maximum and the Peak over threshold data series, respectively. L-moment parameter estimation method and Anderson Darling goodness of fit test were used to identify the best parametric distribution as well as the more suitable data series to represent extreme data. For the non-parametric approach, the kernel density estimation (KDE) is used in this study to determine the best distribution for extreme PM10.

Description

Keywords

Anderson Darling goodness of fit test, generalized extreme value, generalized pareto, kernel density estimation, L-moments, non-parametric distribution, PM10

Citation

Safari, M. A. M., & Zin, W. Z. W. (2017). Pemodelan Taburan Kebarangkalian Zarah Terampai Melampau di Lembah Klang. Sains Malaysiana, 46(6), 989-999.

Collections