Nonlinearity in forecasting energy commodity prices: Evidence from a focused time-delayed neural network

This paper aims to develop an artificial neural networkbased forecasting model employing a nonlinear focused time-delayed network (FTDNN) for energy commodity market forecasts. To validate the proposed model, crude oil and natural gas prices are used period 2007 2020, including Covid-19 period. Empirical findings show that FTDNN outperforms existing baselines models in West Texas Intermediate Brent National Balancing Point Henry Hub prices. As result, we demonstrate predictability of during volatile crisis period, which is attributed flexibility parameters, implying our study can facilitate better understanding dynamics market.