Influence on the Cross-Validated Smoothing Parameter in Spline Smoothing,

Abstract

We address the problem of influence in estimating the smoothing parameter when fitting a univariate smoothing spline. Using the local-influence methods of Cook (1986), a diagnostic is derived to identify observed responses that locally influence the choice of smoothing parameter by generalized cross-validation. The diagnostic motivates a discussion of an apparent sensitivity of generalized cross-validation.

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1992
Accession Number
ADP007133

Entities

People

  • William Thomas

Organizations

  • University of Minnesota

Tags

DTIC Thesaurus Topics

  • Computer Science
  • Data Science
  • Engineering
  • Information Science
  • Sensitivity
  • Statistics
  • Theoretical Computer Science
  • Validation

Fields of Study

  • Mathematics

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Regression Analysis.