Page 376 - Special Topic Session (STS) - Volume 3
P. 376
STS551 Yousif Alyousifi et al.
Empirical bayes method for modelling of air
pollution index
Yousif Alyousifi, Nurulkamal Masseran, Kamarulzaman Ibrahim
Universiti Kebangsaan Malaysia
Abstract
Air pollution is becoming a problem of concern in many parts of the world
nowadays. Monitoring the level of air pollution by using air pollution index
(API) is commonly practiced in Malaysia. In this study, an empirical Bayes
method is applied for estimating the parameters in the Markov transition
probability matrix for describing the stochastic behaviour of API data. The
study reported in this paper is conducted based on the hourly data collected
from the central region in Malaysia for a period of 3 years. The results describe
the experience of air pollution for the region whereby the risk of occurrences
for unhealthy events is small; however, some areas experienced a longer
unhealthy condition and also with a higher probability as compared to the
other areas.
Keywords
Empirical Bayes; Markov Chain Modeling; Air Pollution Index
1. Introduction
The problem of air pollution in Malaysia is an important topic that has
attracted the concern of many researchers (Azid et al.2014; Latif et al 2014;
Masseran et al. 2016; Alyousifi et al 2017; Al-Dhurafi, et al 2018). The urban
and industrial areas in Malaysia are considered to be the most affected due to
the presence of high density of traffic and manufacturing industries (Azid et
al.2014). Based on the studies by (Latif et al. 2014), traffic is known to be one
of the major sources of air pollution in the urban areas for most developing
countries, including Malaysia. In addition, the open burning and forest fires
that often occurred in the neighbouring countries such as Indonesia could be
one source of air pollution in Malaysia.
In Malaysia, since 1998 the Department of Environment (DOE) has adopted
the Air Pollution Index (API) as an indicator of air quality, for providing the
public with information on the quality of air in the environment (DOE 2000;
Masseran et al. 2016). The API value is described by the Department of
Environment as a simple measure for describing the state of air quality in the
environment and providing easily understood information about air pollution.
It is determined based on the maximum value of five sub-indices of pollutants,
namely, particle matter (PM10), sulphur dioxide (SO2), carbon monoxide (CO),
365 | I S I W S C 2 0 1 9