Adaptive covariance estimation of locally stationary processes
Mallat, Stéphane ; Papanicolaou, George ; Zhang, Zhifeng
Ann. Statist., Tome 26 (1998) no. 3, p. 1-47 / Harvested from Project Euclid
It is shown that the covariance operator of a locally stationary process has approximate eigenvectors that are local cosine functions. We model locally stationary processes with pseudo-differential operators that are time-varying convolutions. An adaptive covariance estimation is calculated by searching first for a "best" local cosine basis which approximates the covariance by a band or a diagonal matrix. The estimation is obtained from regularized versions of the diagonal coefficients in the best basis.
Publié le : 1998-02-14
Classification:  Locally stationary processes,  local cosine bases,  adaptive covariance estimation,  approximate Karhunen-Loeve basis,  62M15,  60G15
@article{1030563977,
     author = {Mallat, St\'ephane and Papanicolaou, George and Zhang, Zhifeng},
     title = {Adaptive covariance estimation of locally stationary processes},
     journal = {Ann. Statist.},
     volume = {26},
     number = {3},
     year = {1998},
     pages = { 1-47},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1030563977}
}
Mallat, Stéphane; Papanicolaou, George; Zhang, Zhifeng. Adaptive covariance estimation of locally stationary processes. Ann. Statist., Tome 26 (1998) no. 3, pp.  1-47. http://gdmltest.u-ga.fr/item/1030563977/