Generalized exponential smoothing in prediction of hierarchical time series

Abstract

Shang and Hyndman (2017) proposed a grouped functional time series forecasting approach as a combination of individual forecasts obtained using the generalized least squares method. We modify their methodology using a generalized exponential smoothing technique for the most disaggregated functional time series in order to obtain a more robust predictor. We discuss some properties of our proposals based on the results obtained via simulation studies and analysis of real data related to the prediction of demand for electricity in Australia in 2016.