Forecasting Carbon Monoxide (CO) Pollutant in Putrajaya using the Local Mean Approximation Method

Abstract

Air polution is a great concern to all parties because it is not only affects human health but also affects the environment and the development of the country. This study examines the forecast of carbon monoxide (CO) in metropolitan areas in Malaysia using the local mean approximation method. This study aims to predict the concentration of CO pollutant that has been observed hourly at the selected area which is Putrajaya. The delay time parameter was obtained from the determination method while embedding dimension was obtained from the Cao method. The presence of chaotic dynamics is successfully detected by using the Cao method and the phase space plot. The one-hour forward forecasting process of CO time series using and was carried out using the Local Mean Approximation Method. The result showed that correlation coefficient value is 0.7674 which is close to 1. This forecasting is expected to assist those stakeholders in managing CO pollution effectively especially in metropolitan areas in Malaysia.