Robust Prediction Operations for Stationary Processes.
Abstract
This paper considers prediction for stationary processes, in environments where data outliers may be present. The develops a sequence of outlier resistant prediction operations, which is 'uniformly qualitatively robust. Studied are the asymptotic mean-squared performance of the developed operations, both in the absence and the presence of i.i.d. data outliers. Important performance characteristics studied include the breakdown point and the influence function. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 18, 1987
- Accession Number
- ADA185408
Entities
People
- P. P. Kazakos
Organizations
- University of Virginia