Title :
Analysis of death series by SSA based BSS technique
Author :
Rui Wang;Hong-Guang Ma;Jian-Qiang Qin;Xiao-Wei Feng;Hao-Miao Liu
Author_Institution :
Xi´an Research Institute of High Technology, Xi´an 710025, P.R. China
Abstract :
American death series, which shows the monthly accidental deaths in the USA between 1973 and 1978, is a multicomponent time series. We analyze this well-known series via the singular spectrum analysis (SSA) based blind source separation (BSS) technique. SSA is a powerful approach to decomposing the multicomponent time series. There is an important factor when SSA is used for extracting the principal components from the time series, i.e. a proper window length should be selected for changing the time series into the trajectory matrix. If the window length is not properly selected, the principal components will not be correctly extracted from the time series. In this paper, we propose a method of selecting the window length to reconstruct the trajectory matrix of multicomponent time series, based on which, we conduct the SSA-based BSS test on the death series. The testing results validate the effectiveness of the proposed method.
Keywords :
"Time series analysis","Blind source separation","Trajectory","Matrix decomposition","Correlation","Mathematical model","Covariance matrices"
Conference_Titel :
Information, Communications and Signal Processing (ICICS), 2015 10th International Conference on
DOI :
10.1109/ICICS.2015.7459854