Modelling and forecasting of day-ahead electricity price in Indian energy exchange - evidence from MSARIMA-EGARCH model

Abstract

Electricity prices often exhibit extreme volatility due to its non-storable nature coupled with significant seasonal and diurnal variations of demand, supply constraints at peak hours and transmission bottlenecks. This study tries to forecast day-ahead hourly electricity price of Indian energy exchange (IEX) with an additional objective of modelling the volatility using MSARIMA and MSARIMA-EGARCH models. It has been found that MSARIMA-EGARCH model slightly outperform MSARIMA model in terms of in-sample forecasting performances. The study reveals that seasonality and time-varying volatility are present and past shocks to the variance are asymmetric with negative shocks give rise to higher volatility of price than positive shocks. The study also establishes that shocks to electricity price volatility die out almost instantaneously. The above information can help to build up cost effective risk management plans for the participating companies in IEX.