scientific article; zbMATH DE number 1069597

MaRDI QIDQ4356558zbMATH OpenFDO

Authors Janos Máth, Gy. Terdik

Publication date 1 October 1997

zbMATH Keywords

best linear predictabilitynon-Gaussian

Mathematics Subject Classification ID

Inference from stochastic processes and prediction (62M20)

Review text

We consider cases not necessarily Gaussian when the linear predictor is as good as the linear and quadratic ones together. The assumption of the best linear predictability is concerned with the bispectrum of the innovation series originated from the best linear predictor. The bilinear realizable Hermite degree-2 process with separable kernel is an example of the situation when, although the process is non-Gaussian, the linear predictor is the best among all possible nonlinear ones. For the entire collection see [Zbl 0846.00010].

Recommendations

• A New Test of Linearity of Time Series Based on the Bispectrum
• Bispectral-based goodness-of-fit tests of Gaussianity and linearity of stationary time series
• A TEST FOR NON-LINEARITY OF PREDICTION IN TIME SERIES
• scientific article; zbMATH DE number 3854250
• scientific article; zbMATH DE number 718746
• scientific article; zbMATH DE number 4157734
• Reliable detection of nonlinearity in experimental time series with strong periodic components
• Testing for time series linearity

Cited in

(3)

• scientific article; zbMATH DE number 1336728 (Why is no real title available?)
• scientific article; zbMATH DE number 957396 (Why is no real title available?)
• A RELATION FOR ‘LINEARITY’ OF THE BISPECTRUM

This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4356558)