Covariance matrix analysis for higher order fractional Brownian motion time series
Department or Administrative Unit
Fractional Brownian motion (fBm) is an important mathematical model for describing a range of phenomena and processes. The properties of discrete time fBm (dfBm) when m equals 1 and 2 have been reported in the literature. This paper focuses on analysis of auto-covariance matrix of the m-th order (m > 2) of a dfBm process and the error associated with the approximation of a large dimensional auto-covariance matrix. Applying matrix theory and analysis, we also generalize the asymptotic properties of the eigenvalues of the auto-covariance matrix. Based on the analysis, two theorems and one lemma are proposed and their proofs are provided. Your goal is to simulate, as closely as possible, the usual appearance of typeset papers. This document provides an example of the desired layout and contains information regarding desktop publishing format, type sizes, and type faces.
Montillet, J.-P., & Yu, K. (2015). Covariance matrix analysis for higher order fractional Brownian motion time series. 2015 IEEE 28th Canadian Conference on Electrical and Computer Engineering (CCECE). https://doi.org/10.1109/ccece.2015.7129488
2015 IEEE 28th Canadian Conference on Electrical and Computer Engineering (CCECE)
Copyright © 2015, IEEE