A hierarchical approach to covariance function estimation for time series

Abstract

The covariance function in time series models is typically modeled via a parametric family. This ensures straightforward linear prediction while maintaining positive-de niteness of the covariance function. We suggest an alternative approach, which will result in data-determined shrinkage towards this parametric model. Positive-de niteness is maintained by… (More)

Topics

1 Figure or Table

Cite this paper

@inproceedings{Daniels2000AHA, title={A hierarchical approach to covariance function estimation for time series}, author={Michael J. Daniels}, year={2000} }