On the predictability of energy commodity markets by an entropy-based computational method

This paper proposes a novel computational method for assessing the predictability of commodity market time series, by predicting the entropy of the series under investigation. Assessing the predictability of a time series is the first mandatory step in order to further apply low-risk and efficient price forecasting methods. According to conventional entropy-based analysis (where the entropy is always ex-post estimated), high entropy values characterize unpredictable series, while more stable series exhibits lesser entropy values. Here, we predict (i.e. ex-ante) the entropy regarding the future behavior of a series, based on the observation of historical data. Our prediction is performed according to the optimum least squares minimization algorithm, usually used in many computational aspects of management science. Preliminary results, applied to energy commodity futures, show the effectiveness of the proposed method for application to energy market time series.