Functional Time Series Forecasting: Functional Singular Spectrum Analysis Approaches
This work provides improved forecasting techniques for functional time series, particularly beneficial for researchers and practitioners dealing with periodic stochastic processes.
This paper introduces two nonparametric methods, functional singular spectrum analysis recurrent forecasting and vector forecasting, for predicting functional time-dependent data. The methods are shown to outperform a gold standard algorithm for periodic stochastic processes.
In this paper, we propose two nonparametric methods used in the forecasting of functional time-dependent data, namely functional singular spectrum analysis recurrent forecasting and vector forecasting. Both algorithms utilize the results of functional singular spectrum analysis and past observations in order to predict future data points where recurrent forecasting predicts one function at a time and the vector forecasting makes predictions using functional vectors. We compare our forecasting methods to a gold standard algorithm used in the prediction of functional, time-dependent data by way of simulation and real data and we find our techniques do better for periodic stochastic processes.