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 generalized least squares method. We modify their methodology using generalized exponential smoothing technique for the most disaggregated functional time series in order to obtain more robust predictor. We discuss some properties of our proposals basing on results obtained via simulation studies and analysis of real data related to a prediction of a demand for electricity in Australia in 2016.

Aggregated moving functional median in robust prediction of hierarchical functional time series - an application to forecasting web portal users behaviors

“…[35]) which take into account the complete hierarchy. The method has been recently modified using a generalized exponential smoothing technique for the most disaggregated functional time series in order to obtain a more robust predictor (for details see [21]). Shang and Hyndman's method of mortality rates forecasting uses a functional principal component analysis in order to perform functional principal component regression at each level, where the times series forecast of the principal component scores are obtained with an effective but nonrobust univariate time series forecasting method, namely Hyndman and Shang [11] method (see also [15]).…”