Regression and multivariate models for predicting particulate matter concentration level

The devastating health effects of particulate matter (PM10) exposure by susceptible populace has made it necessary to evaluate PM10 pollution. Meteorological parameters and seasonal variation increases PM10 concentration levels, especially in areas that have multiple anthropogenic activities. Hence,...

Full description

Main Authors: Nazif, A., Mohammed, N.I., Malakahmad, A., Abualqumboz, M.S.
Format: Article
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-eprints.21908 /
Published: Springer Verlag 2018
Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85031419991&doi=10.1007%2fs11356-017-0407-2&partnerID=40&md5=2ebc6364619c0d150844c58e03878e8b
http://eprints.utp.edu.my/21908/
Tags: Add Tag
No Tags, Be the first to tag this record!
id utp-eprints.21908
recordtype eprints
spelling utp-eprints.219082018-08-01T01:13:21Z Regression and multivariate models for predicting particulate matter concentration level Nazif, A. Mohammed, N.I. Malakahmad, A. Abualqumboz, M.S. The devastating health effects of particulate matter (PM10) exposure by susceptible populace has made it necessary to evaluate PM10 pollution. Meteorological parameters and seasonal variation increases PM10 concentration levels, especially in areas that have multiple anthropogenic activities. Hence, stepwise regression (SR), multiple linear regression (MLR) and principal component regression (PCR) analyses were used to analyse daily average PM10 concentration levels. The analyses were carried out using daily average PM10 concentration, temperature, humidity, wind speed and wind direction data from 2006 to 2010. The data was from an industrial air quality monitoring station in Malaysia. The SR analysis established that meteorological parameters had less influence on PM10 concentration levels having coefficient of determination (R2) result from 23 to 29 based on seasoned and unseasoned analysis. While, the result of the prediction analysis showed that PCR models had a better R2 result than MLR methods. The results for the analyses based on both seasoned and unseasoned data established that MLR models had R2 result from 0.50 to 0.60. While, PCR models had R2 result from 0.66 to 0.89. In addition, the validation analysis using 2016 data also recognised that the PCR model outperformed the MLR model, with the PCR model for the seasoned analysis having the best result. These analyses will aid in achieving sustainable air quality management strategies. © 2017, Springer-Verlag GmbH Germany. Springer Verlag 2018 Article NonPeerReviewed https://www.scopus.com/inward/record.uri?eid=2-s2.0-85031419991&doi=10.1007%2fs11356-017-0407-2&partnerID=40&md5=2ebc6364619c0d150844c58e03878e8b Nazif, A. and Mohammed, N.I. and Malakahmad, A. and Abualqumboz, M.S. (2018) Regression and multivariate models for predicting particulate matter concentration level. Environmental Science and Pollution Research, 25 (1). pp. 283-289. http://eprints.utp.edu.my/21908/
institution Universiti Teknologi Petronas
collection UTP Institutional Repository
description The devastating health effects of particulate matter (PM10) exposure by susceptible populace has made it necessary to evaluate PM10 pollution. Meteorological parameters and seasonal variation increases PM10 concentration levels, especially in areas that have multiple anthropogenic activities. Hence, stepwise regression (SR), multiple linear regression (MLR) and principal component regression (PCR) analyses were used to analyse daily average PM10 concentration levels. The analyses were carried out using daily average PM10 concentration, temperature, humidity, wind speed and wind direction data from 2006 to 2010. The data was from an industrial air quality monitoring station in Malaysia. The SR analysis established that meteorological parameters had less influence on PM10 concentration levels having coefficient of determination (R2) result from 23 to 29 based on seasoned and unseasoned analysis. While, the result of the prediction analysis showed that PCR models had a better R2 result than MLR methods. The results for the analyses based on both seasoned and unseasoned data established that MLR models had R2 result from 0.50 to 0.60. While, PCR models had R2 result from 0.66 to 0.89. In addition, the validation analysis using 2016 data also recognised that the PCR model outperformed the MLR model, with the PCR model for the seasoned analysis having the best result. These analyses will aid in achieving sustainable air quality management strategies. © 2017, Springer-Verlag GmbH Germany.
format Article
author Nazif, A.
Mohammed, N.I.
Malakahmad, A.
Abualqumboz, M.S.
spellingShingle Nazif, A.
Mohammed, N.I.
Malakahmad, A.
Abualqumboz, M.S.
Regression and multivariate models for predicting particulate matter concentration level
author_sort Nazif, A.
title Regression and multivariate models for predicting particulate matter concentration level
title_short Regression and multivariate models for predicting particulate matter concentration level
title_full Regression and multivariate models for predicting particulate matter concentration level
title_fullStr Regression and multivariate models for predicting particulate matter concentration level
title_full_unstemmed Regression and multivariate models for predicting particulate matter concentration level
title_sort regression and multivariate models for predicting particulate matter concentration level
publisher Springer Verlag
publishDate 2018
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-85031419991&doi=10.1007%2fs11356-017-0407-2&partnerID=40&md5=2ebc6364619c0d150844c58e03878e8b
http://eprints.utp.edu.my/21908/
_version_ 1741196540363407360
score 11.62408

Similar Items