New Improved Chaotic Approach Model Application on Forecasting Ozone Concentration Time Series

This study is an application of chaotic approach on forecasting the ozone air pollutant time series at Malaysian background station located in Jerantut, Pahang. Before the forecasting model can be built, the time series are tested in advance whether the nature is chaotic or not. Through phase space plot and Cao method, the ozone air pollutant time series were found to be low in dimensional chaotic. Therefore, the forecasting model through local linear approximation is constructed. As an innovation, this model is improved. As comparison, the linear regression forecasting model was also constructed. By calculating the mean absolute error, root mean square error and correlation coefficient, the results showed that the new improved local linear approximation model is better than the other models. Thus, the improvement was worth it. Therefore, chaotic approach is an alternative approach that can be used to construct forecasting model for ozone pollutants time series. The discovery of new method in this study is expected to help facilitate the efforts of stakeholders in dealing with the issues of air pollution, especially ozone.