Abstract:
Determination of embedding dimension based on symplectic geometry. Let X be lag matrix of y, A=X'X, and A(n) is (n-1) times transformed form of A by Householder matrix. If descending sorted eigenvalues A(n) tend to be constant for dimension d+1 then d is the proper embedding dimension. let sigma be square of eigenvalues of A matrix, sigma1=(lambda^2)max,…,sigman=(lambda^2)min,where n is the dimension of X. If sigma1>sigma2>…>sigmad>sigma(d+1)>=...>=sigman, then d is the embedding dimension of the reconstruction system(Lei et al. 2002). The dimension before horizontal section of log(sigmai/tr(sigmai))plot is the proper dimension.
More software in Statistical Software Components from Boston College Department of Economics Address: Boston College, 140 Commonwealth Avenue, Chestnut Hill MA 02467 USA Contact information at EDIRC. Series data maintained by Christopher F Baum ().
This site is part of RePEc
and all the data displayed here is part of the RePEc data set.
Is your work missing from RePEc? Here is how to
contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .