Predicting Carbon Monoxide Time Series Between Different Settlements Area in Malaysia Through Chaotic Approach

Abstract

This research is designed to analysing and predicting the time series of Carbon Monoxide (CO) at two stations located in Petaling Jaya and Jerantut through chaotic approach. Both stations were selected to represent CO time series in the urban and rural settlements. Chaotic approach has two steps, which are to (i) detect the presence of chaotic dynamics through phase space plot and Cao method and (ii) predict the time series which is done using local linear approximation method (LLAM). Through step (i), the result from phase space plot and parameter of Cao method shown that the presence of chaotic dynamics has been detected. Therefore, the chaotic approach is used to predict the time series. While in step (ii), the results shown that the correlation coefficient value of LLAM are 0.7536 for Petaling Jaya (τ=1, m=4) and 0.6946 for Jerantut (τ=1, m=5) which are close to one. This shows a positive sign that chaotic approach is applicable in both types of urban and rural settlements areas. These finding are expected to help stakeholders such as Ministry of Education and Department of Environment to having a better air pollution management.