Department of Economics and Business Economics

Asymptotic Theory of Outlier Detection Algorithms for Linear Time Series Regression Models

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

  • Søren Johansen
  • Bent Nielsen, University of Oxford - Nuffield College, United Kingdom

Outlier detection algorithms are intimately connected with robust statistics that down-weight some observations to zero. We define a number of outlier detection algorithms related to the Huber-skip and least trimmed squares estimators, including the one-step Huber-skip estimator and the forward search. Next, we review a recently developed asymptotic theory of these. Finally, we analyse the gauge, the fraction of wrongly detected outliers, for a number of outlier detection algorithms and establish an asymptotic normal and a Poisson theory for the gauge.

Original languageEnglish
JournalScandinavian Journal of Statistics
Pages (from-to)321-348
Number of pages28
Publication statusPublished - 1 Jun 2016

    Research areas

  • Forward search, gauge, Huber-skip, impulse indicator saturation, iterated martingale inequality, iteration of one-step estimators, one-step Huber-skip, robustified least squares, weighted and marked empirical processes

See relations at Aarhus University Citationformats

Download statistics

No data available

ID: 101193775