A Hybrid ARIMA and Neural Networks Model for PM-10 Pollution Estimation: The Case of Chiang Mai City Moat Area

Most of time series model are usually investigated and implemented by ARIMA and Neural Networks (NNs) model. However, ARIMA model may not be adequate for complex patterned problem while NNs model can well reveal the correlation of nonlinear pattern. Since, over-fitting due to a learning process is the main advantage of NNs as well as local trapped of parameters due to the large structure of the networks. To improve the forecast performance of both ARIMA and NNs for high accuracy, hybrid ARIMA and NNs model is alternate selected and employed to examine the Chiangmai city moat's PM-10 time series data. The experimental results demonstrated that the hybrid model outperformed best over NNs and ARIMA respectively.