UPM Institutional Repository

Bivariate extreme value with application to PM10 concentration analysis


Mohd Amin, Nor Azrita and Adam, Mohd Bakri and Ibrahim, Noor Akma and Aris, Ahmad Zaharin (2014) Bivariate extreme value with application to PM10 concentration analysis. In: International Conference on Mathematics, Engineering and Industrial Applications 2014 (ICoMEIA 2014), 28-30 May 2014, Penang, Malaysia. .


This study is focus on a bivariate extreme of renormalized componentwise maxima with generalized extreme value distribution as a marginal function. The limiting joint distribution of several parametric models are presented. Maximum likelihood estimation is employed for parameter estimations and the best model is selected based on the Akaike Information Criterion. The weekly and monthly componentwise maxima series are extracted from the original observations of daily maxima PM10 data for two air quality monitoring stations located in Pasir Gudang and Johor Bahru. The 10 years data are considered for both stations from year 2001 to 2010. The asymmetric negative logistic model is found as the best fit bivariate extreme model for both weekly and monthly maxima componentwise series. However the dependence parameters show that the variables for weekly maxima series is more dependence to each other compared to the monthly maxima.

Download File

PDF (Abstract)
Bivariate extreme value with application to PM10 concentration analysis.pdf

Download (35kB) | Preview

Additional Metadata

Item Type: Conference or Workshop Item (Paper)
Divisions: Faculty of Environmental Studies
Institute for Mathematical Research
DOI Number: https://doi.org/10.1063/1.4915672
Publisher: AIP Publishing LLC
Keywords: Air quality; Akaike information criterion; Bivariate extreme; Maximum likelihood estimation; Parametric models
Depositing User: Nabilah Mustapa
Date Deposited: 26 Sep 2017 04:06
Last Modified: 26 Sep 2017 04:06
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.1063/1.4915672
URI: http://psasir.upm.edu.my/id/eprint/57345
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item