A Sieve Estimator for the Covariance of a Gaussian Process
Beder, Jay H.
Ann. Statist., Tome 16 (1988) no. 1, p. 648-660 / Harvested from Project Euclid
Maximum likelihood estimation for the covariance $R$ of a zero-mean Gaussian process is considered, with no assumptions on the covariance or the "time" parameter set $T$. It is shown that the likelihood function is a.s. unbounded in general, and a sieve estimator $\hat{R}$ is constructed. The distribution of $\hat{R}$, considered as a process on $T \times T$, can be described exactly if a certain technical assumption is satisfied concerning the bivariate series expansion of $R$. It is then shown that $\hat{R}(s, t)$ is asymptotically unbiased and consistent (weakly and in mean square) at each $(s, t) \in T \times T$, and that $\hat{R}$ is strongly consistent (globally) in an appropriate norm.
Publié le : 1988-06-14
Classification:  Consistency,  Gaussian dichotomy theorem,  maximum likelihood estimation,  reproducing kernel Hilbert space,  sieve,  62M09,  60G15,  60G30
@article{1176350825,
     author = {Beder, Jay H.},
     title = {A Sieve Estimator for the Covariance of a Gaussian Process},
     journal = {Ann. Statist.},
     volume = {16},
     number = {1},
     year = {1988},
     pages = { 648-660},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176350825}
}
Beder, Jay H. A Sieve Estimator for the Covariance of a Gaussian Process. Ann. Statist., Tome 16 (1988) no. 1, pp.  648-660. http://gdmltest.u-ga.fr/item/1176350825/