Title :
Sequential learning of differential trend
Author_Institution :
Inst. for Infocomm Res., A*STAR (Agency for Sci., Technol. & Res.), Singapore, Singapore
Abstract :
This paper describes a simple learning method to sequentially select recent time series values as features to model the differential trend of a time series. This method is used to solve the First International Competition on Time Series Forecasting (ICTSF) [1]. The objective of ICTSF is to predict eight time series with different time frequency and different forecasting horizon. Experimental result shows viability of the method in multi-step forecasting.
Keywords :
forecasting theory; learning (artificial intelligence); time series; First International Competition on Time Series Forecasting; differential trend modeling; forecasting horizon; lCTSF; multistep forecasting; sequential learning; time frequency; time series sequential selection; Bayesian methods; Electronic learning; Predictive models; Robots;
Conference_Titel :
Evolving and Adaptive Intelligent Systems (EAIS), 2012 IEEE Conference on
Conference_Location :
Madrid
Print_ISBN :
978-1-4673-1728-3
Electronic_ISBN :
978-1-4673-1726-9
DOI :
10.1109/EAIS.2012.6232830
