Statistical Model Approach of Electricity Price Forecasting for Indian Electricity Market

The cost of renewable power price is coming down with increased modest methods in the electricity market. When it is to benefit both the producers and consumers of electricity, a next-day electricity price forecasting technique is exceedingly welcomed. The research work focuses on the comparison of two forecasting methods applied in the Indian electricity market scenario which is operating on a day-ahead basis. In this market, the electricity price series is less volatile and highly correlated to forecast electricity prices with better accuracy. The Seasonal Auto-Regressive Integrated Moving Average (SARIMA) and Generalized Auto-Regressive Conditional Heteroskedastic (GARCH) models are employed on the time series data to predict and forecast day-ahead prices. The price data from the real-time market is taken from the Indian Energy Exchange (IEX) site. The Mean Absolute Percentage Error (MAPE) and the Root Mean Squared Error (RMSE) tests are also presented to check the robustness of the model. The modelling of the systems is done in the Jupyter Notebook using Python.