پيش بيني آلودگي هواي شيراز

Time Series Model of Air Pollution Forecast in Shiraz

Today, air pollution is one of the main problems of most cities. People spread a large amount of various materials which is inconsistent with natural mechanisms, in the air every day. With consideration of programming, using proper techniques of prediction is one of the effective aims of organization. Due to the importance of air pollution in human health, forecasting of the pollution index can be useful for futures deciding. In the last 10 years, many advances have been made in the statistical modeling of air pollution time-series studies. In this study, time series of pollution standard index (PSI) was obtained from data which were gathered from the Shiraz pollution measuring station of air pollution. PSI was then used for forecasting by regression and non regression methods. ARMA (3,3) model was selected as the best model for forecasting according to Mean Absolute Percentage Error (MAPE) criteria. The procedure for finding an adequate ARMA (p.q) model to represent the series can be described as an iterative process involving three key steps: 1) Identification of model to be used, 2) estimation of the parameters of the model, and 3) checking the candidate model by appropriate criteria. The results indicate that this model can predict air pollution with high accuracy