Application of Functional Time Series Model in Forecasting Monthly Diurnal API Curves: A Comparison between Multi-Step Ahead and Iterative One-Step Ahead Approach

In Malaysia, Air Pollution Index (API) is used to assess the status of background air quality. The computation API involved six major pollutants including PM10, PM2.5, O3, CO, SO2 and NOx. Due harmful effect pollution, forecasting important. This paper introduces application Functional Time Series (FTS) model in monthly diurnal maximum curves at two selected sites Peninsular Malaysia; namely Shah Alam Selangor Pasir Gudang Johor. Two FTS models were compared which include Multi-Step ahead Iterative One-Step approach. results show that has produced better performance giving lowest error measures; FMSE, FRMSE FMAPE ahead. study shown advantage because it enables prediction continuous levels within a defined continuum time, this was interval time 24 hours. descriptive mean shows bimodal pattern with peak 3.00 pm average are healthy level. exhibits an increasing after sunrise towards 10.00 am both sites, inform that, higher contribution vehicles emission while standard deviation differs pattern. recommended as alternative be by government environmentalists providing input for guiding pollution control protecting public health early stage. Furthermore, private sector industries, might provide predictive analytic tool daily instead single value.