Community Research and Development Information Service - CORDIS

Abstract

The DoD method for robust estimating of variances of measurement series (time series) without the exclusion of outliers is described in detail, including its mathematical-statistical basis. In particular, it is proved that the total of all n(n-1)/2 absolute differences obtainable from n measurement values (realisations of random variables) may be subdivided into groups, each of them consisting of stochastically independent differences only. This is necessary for theoretical foundation of the DoDM estimator. Numerical examples of possible applications are given. This completes the evaluation of earlier IDA-80 data by applying the DoDM estimator.

Additional information

Authors: BEYRICH W, Institut für Datenverarbeitung in der Technik, Kernforschungszentrum Karlsruhe GmbH, Karlsruhe (DE);GOLLY W, Institut für Datenverarbeitung in der Technik, Kernforschungszentrum Karlsruhe GmbH, Karlsruhe (DE);PETER N, Institut für Datenverarbeitung in der Technik, Kernforschungszentrum Karlsruhe GmbH, Karlsruhe (DE);SEIFERT R, Institut für Datenverarbeitung in der Technik, Kernforschungszentrum Karlsruhe GmbH, Karlsruhe (DE)
Bibliographic Reference: EUR 11398 EN (1990) KfK 4721, 50 pp., MF, ECU 4
Availability: (2)
Record Number: 199011084 / Last updated on: 1994-12-01
Category: PUBLICATION
Original language: en
Available languages: en
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top