Adaptive estimation of stationary Gaussian fields

Details Adaptive estimation of stationary Gaussian fields Adaptive estimation of stationary Gaussian fields

2009-01-15

arxiv.org/abs/0901.2212v2

Mathematics

Statistics Theory

Scientific paper

We study the nonparametric covariance estimation of a stationary Gaussian field X observed on a regular lattice. In the time series setting, some procedures like AIC are proved to achieve optimal model selection among autoregressive models. However, there exists no such equivalent results of adaptivity in a spatial setting. By considering collections of Gaussian Markov random fields (GMRF) as approximation sets for the distribution of X, we introduce a novel model selection procedure for spatial fields. For all neighborhoods m in a given collection M, this procedure first amounts to computing a covariance estimator of X within the GMRFs of neighborhood m. Then, it selects a neighborhood by applying a penalization strategy. The so-defined method satisfies a nonasymptotic oracle type inequality. If X is a GMRF, the procedure is also minimax adaptive to the sparsity of its neighborhood. More generally, the procedure is adaptive to the rate of approximation of the true distribution by GMRFs with growing neighborhoods.

Affiliated with

Also associated with

No associations

Rating

Adaptive estimation of stationary Gaussian fields does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Adaptive estimation of stationary Gaussian fields, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive estimation of stationary Gaussian fields will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-589735