A Novel Hybrid Algorithm to Forecast Functional Time Series Based on Pattern Sequence Similarity with Application to Electricity Demand

Abstract

The forecasting of future values is a very challenging task. In almost all scientific disciplines, the analysis of time series provides useful information and even economic benefits. In this context, this paper proposes a novel hybrid algorithm to forecast functional time series with arbitrary prediction horizons. It integrates a well-known clustering functional data algorithm into a forecasting strategy based on pattern sequence similarity, which was originally developed for discrete time series. The new approach assumes that some patterns are repeated over time, and it attempts to discover them and evaluate their immediate future. Hence, the algorithm first applies a clustering functional time series algorithm, i.e., it assigns labels to every data unit (it may represent either one hour, or one day, or any arbitrary length). As a result, the time series is transformed into a sequence of labels. Later, it retrieves the sequence of labels occurring just after the sample that we want to be forecasted. This sequence is searched for within the historical data, and every time it is found, the sample immediately after is stored. Once the searching process is terminated, the output is generated by weighting all stored data. The performance of the approach has been tested on real-world datasets related to electricity demand and compared to other existing methods, reporting very promising results. Finally, a statistical significance test has been carried out to confirm the suitability of the election of the compared methods. In conclusion, a novel algorithm to forecast functional time series is proposed with very satisfactory results when assessed in the context of electricity demand.

A Novel Ensemble Method for Electric Vehicle Power Consumption Forecasting: Application to the Spanish System

“…For the ARIMA and the GARCH algorithms, the parameters of the corresponding models have been calculated by using a one-year length electric vehicle power consumption observations. Given the patterns observed in the weekly time series, two models have been defined for each algorithm: one for Sundays and Mondays (SM models); and one for Tuesdays to Saturdays (TS models), which is also a common strategy in this scenario [41]. This implies that the differentiated time series is split into two subseries, one containing all Sundays and Mondays to calculate the SM models; and another containing the rest of the days for calculating the TS models.…”

Dynamics of the Covid-19 Pandemics: Global Pattern and Between Countries Variations

“…The study of time series is a very active field of research, this data format frequent and exploited, especially in acoustics [6], power management [7] or health [8]. In addition to prediction, time series can also be considered from a classification point of view, i.e.…”

Modeling of Compressive Strength of Self-Compacting Rubberized Concrete Using Machine Learning

“…Therefore, in this article, an effort was made to model the compressive strength of SCRC by adopting one of several machine learning methods. So far, metaheuristic methods, and especially neural networks, have been successfully applied in various fields, such as in the control and optimization of processes, economics, medicine, and engineering [ 6 , 7 , 8 , 9 , 10 ]. They have also been used to model the properties of concrete in fresh or solid state [ 11 , 12 , 13 , 14 , 15 ], but much less in concrete with the addition of rubber [ 16 , 17 , 18 ].…”