INPUT VARIABLE SELECTION FOR HOURLY OZONE (O3) CONCENTRATION PREDICTION IN URBAN AREA

• Main Authors: Napi N.N.L.M., Abdullah S., Mansor A.A., Ahmed A.N., Mansor W.N.W., Ismail M.
• Other Authors: 57224902975
• Format: Article
• Published: Asian Research Publishing Network 2023

The higher emission of the air pollutants from motor vehicles in urban areas increases the formation of ozone in the atmosphere and give a negative impacts on human health, especially. The multicollinearity becomes a problem as many parameters were considered in the development of air pollutant prediction models. The aimed of the study is to develop the best O3,t+1 prediction model in Terengganu�s urban areas to help local authorities by providing early information about the air quality level. The air pollutants and meteorological data from the year 2010 to 2013 were obtained from the Department of Environment, Malaysia. There are three different models were developed based on three different input selection method. MLR3 had highest correlation coefficient R2, 0.411 compared with MLR1 (0.382) and MLR2 (0.399). Hence, it was selected as the best-fitted model to predict the O3 concentration in Terengganu�s urban areas. �2006-2021 Asian Research Publishing Network (ARPN). All rights reserved.