Forecasting volatility in the EPEX market

This paper uses high-frequency intraday electricity prices from the EPEX market to estimate and forecast realised volatility. Variation is broken down into jump and continuous components using quadratic variation theory. Then several heterogeneous autoregressive models are estimated for the logarithmic and standard deviation transformations. GARCH structures are included in the error terms of the models when evidence of conditional heteroscedasticity is found. Model selection is based on various out-of-sample criteria. Under the logarithmic transformation the simplest model outperforms the rest. Under the standard deviation transformation, jump detection before model estimation is useful to improve forecasting.