A Fourth Order Deconvolution Technique for Nongaussian Linear Processes.

Abstract

In Lii and Rosenblatt (1982) a deconvolution scheme for nonGaussian linear processes making use of third order moments (or spectra) was presented. This is appropriate for such processes with nonzero third order central moments. However, if the third order moments are zero (this could happen in the case of symmetric distributions) it is appropriate to look for a fourth order technique that would be effective. Such a scheme is presented and discussed in this paper together with some illustrative examples.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1982
Accession Number
ADA120663

Entities

People

  • Keh-shin Lii
  • M. Rosenblatt

Organizations

  • University of California, San Diego

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Bandwidth
  • California
  • Celestial Brightness
  • Coefficients
  • Computations
  • Covariance
  • Estimators
  • Extrapolation
  • Integrals
  • Mathematics
  • Military Research
  • Observation
  • Probability
  • Random Variables
  • Sequences
  • Spectra
  • Transfer Functions

Readers

  • Approximation Theory.
  • Statistical inference.