Asymptotic correct correlation tests in model validation
1993 (English)In: Proceedings of the IEEE Conference on Decision and Control, San Antonio, TX, USA, 1993, Vol. 3, no Piscataway, NJ, United States, 2058-2059 p.Conference paper (Refereed)
It is well-known that correlation tests may give true significance levels that differ significantly from the desired ones; the tests are less inclined to reject the null hypothesis when second-hand data are used compared with how they are designed to behave and the situation is the opposite for 'fresh' data. The reason is that the tests are based on the assumption that the limit model (corresponding to infinite data) is available. In this contribution we propose a methodology to design correlation tests that avoid this artifact. This leads to tests of higher order correlations.
Place, publisher, year, edition, pages
San Antonio, TX, USA, 1993. Vol. 3, no Piscataway, NJ, United States, 2058-2059 p.
, Proceedings of the 32nd IEEE Conference on Decision and Control. Part 3 (of 4)
Control theory, Correlation methods, Mathematical models, Matrix algebra, Parameter estimation, Correlation tests, Fresh data, Model validation, Identification (control systems)
Research subject SRA - ICT
IdentifiersURN: urn:nbn:se:kth:diva-60595DOI: 10.1109/CDC.1993.325560OAI: oai:DiVA.org:kth-60595DiVA: diva2:477504
Sponsors: IEEE Control Systems Society NR 201408052012-01-132012-01-132012-01-13Bibliographically approved