No label defined (Q3976176)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 18763
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 18763

      Statements

      0 references
      0 references
      26 June 1992
      0 references
      influence function
      0 references
      breakdown point
      0 references
      robust scale estimates
      0 references
      BIF- estimator
      0 references
      robust filtering
      0 references
      S-estimates
      0 references
      prediction residuals
      0 references
      robust predictor
      0 references
      AR(1)-process
      0 references
      bounded influence autoregression estimator
      0 references
      simulated maximum bias curve
      0 references
      point-mass contamination
      0 references
      new class of robust estimates of autoregression parameters
      0 references
      [For the entire collection see Zbl 0731.00018.] A new class of robust estimates of autoregression parameters is proposed the essential feature of which is the combined use of robust filtering and S-estimates for regression. These robust estimates minimize a highly robust scale measure for observation prediction residuals based on a robust predictor. For the case of the AR(1)-processes the authors compare their new procedure with two other estimators, the Chang, Tiao and Chen estimator and a bounded influence (BIF) autoregression estimator. The comparison is based on the simulated maximum bias curve for point-mass contamination. In this sense the performance of the new estimator dominates that of both, the \textit{I. Chang}, \textit{G. C. Tiao} and \textit{C. Chen} [Technometrics 30, No. 2, 193-204 (1988)] and the BIF estimator, while the BIF estimator dominates the Chang, Tiao and Chen estimator.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references