Robust Estimation of Reduced Rank Models to Large Spatial Datasets

For large datasets, spatial covariances are often modeled using basis functions and covariance of a reduced dimensional latent spatial process. For skewed data, likelihood based approaches with Gaussian assumption may not lead to faithful inference. Any L2 norm based estimation is susceptible to lo...

Full description

Bibliographic Details
Main Authors: Casey M. Jelsema, Rajib Paul, Joseph W. McKean
Format: Article
Language:English
Published: Instituto Nacional de Estatística | Statistics Portugal 2020-05-01
Series:Revstat Statistical Journal
Subjects:
Online Access:https://revstat.ine.pt/index.php/REVSTAT/article/view/297