Abstract—Electricity is one of the most important sources of energy on earth. Today, electricity has become a part of our life. Electricity is the key component to modern technology and without it most of the products that we use simply could not work. Without doubt, the economic growth for almost every country in the world is affected by electricity rates. Therefore, the prediction of electricity sales is very important for Taiwanese economy. This study employs the autoregressive integrated moving average (ARIMA), artificial neural networks (ANN) and the integrated ARIMA-ANN approaches for predicting the industrial electricity and commercial electricity sales (IECES) in Taiwan. The forecasting accuracy measure is based on the mean absolute percentage error. The real dataset, from the years 2006 to 2016, for IECES in Taiwan are collected and analyzed. The prediction results show that the ARIMA-ANN model has the most satisfactory forecasting accuracy for predictions of IECES in Taiwan.

Index Terms—Prediction, electricity sales, ARIMA, artificial neural networks.

The authors are with Department of Statistics and Information Science, Fu Jen Catholic University, New Taipei City, Taiwan, R.O.C. (email: stat1003@mail.fju.edu.tw)

Cite: Yuehjen E. Shao, Yi-Shan Tsai, "Predictions of Industrial and Commercial Electricity Sales in Taiwan Using ARIMA and Artificial Neural Networks Techniques," International Journal of e-Education, e-Business, e-Management and e-Learning vol. 8, no. 2, pp. 74-81, 2018.