1Department of Mathematics, National Institute of Technology Calicut, India
Journal of Mathematical Sciences and Applications.
2017,
Vol. 5 No. 1, 1-16
DOI: 10.12691/jmsa-5-1-1
Copyright © 2017 Science and Education PublishingCite this paper: Chithralekha. K, Jessy John C. Modelling of Multivariate Stationary Time Series Using Rational Approximation of the Spectral Density Function and Wavelet Coherence.
Journal of Mathematical Sciences and Applications. 2017; 5(1):1-16. doi: 10.12691/jmsa-5-1-1.
Correspondence to: Chithralekha. K, Department of Mathematics, National Institute of Technology Calicut, India. Email:
chithrakc@gmail.comAbstract
This paper presents a new method for the modelling of multivariate stationary time series by applying multiple input-single output transfer function noise model, rational approximation of spectral density function and wavelet coherence. Parameter estimation process is simple and the number of parameters needs to be estimated is very less, is the main advantage of this method. The method is verified by simulation studies and it is also applied to model US hog data with five component series.
Keywords