Using ARMAX-EGARCH Model to Forecast Day-ahead Electricity Prices for PJM Market

Under deregulated environment, accurate price forecasting provides crucial information for electricity market participants to make reasonable competing strategies. Three time series models, namely ARMAX, ARMAX-EGARCH and ARMAX-EGARCH-M, for day-ahead electricity prices of the PJM electricity market are proposed, in which the effect of load on the day-ahead electricity prices can be taken into account. The in- and out-of-sample forecasting performance of the respective models are examined. The results from ARMAX models reveal the presence of ARCH. The asymmetric time-varying volatility is studied by ARMAX-EGARCH models and the EGARCH specification for the variance equation demonstrates the absence of a positive leverage effect in day-ahead electricity prices for the PJM electricity market. With respect to forecasts, the ARMAX-EGARCH-M model outperforms the others in terms of the in- and out-of-sample forecasting performance based on four forecast evaluation statistics.