Day ahead electricity price forecast by NARX model with LASSO based features selection

The availability of accurate day-ahead price forecasts is crucial to achieve an effective participation to liberalized electricity markets. Starting from available state of the art, we propose a forecast technique exploiting a nonlinear-ARX model including a feature regression mechanism based on LASSO. The rational behind such choice is twofold. On the one hand, we aim to target potential increase of forecast accuracy by learning complex non-linear mappings. On the other hand, the foremost challenges regard the increase of model interpretability and the minimization of the effort needed to properly set up the forecaster. A framework capable to self-extract features from spot price multi-variate time series might represent a very useful tool for industrial practitioners. Experiments have been performed on Italian market dataset, demonstrating the capabilities of proposed method in extracting useful features and achieving robust performances. Moreover, we show how the proposed method can support interpretation of forecaster structure and reveal interesting cross-correlations within the regression set.