Studying Error Propagation for Energy Forecasting Using Univariate and Multivariate Machine Learning Algorithms

摘要

Statistical machine learning models are widely used in time series forecasting. These models often use historical data recursively to make predictions, i.e. future timesteps. This leads to compounding of errors, which may negatively impact the prediction accuracy for long-term prediction tasks. In this paper, we address this problem by using features that can have “anchoring” effect on recurrent forecasts, thus, limiting the impact of compounding errors. The approach is tested with four machine learning models applied to a benchmark energy dataset. It is observed that the addition of generated features improves performance for both short and long time horizons.