Properties and generation of non-Gaussian time series
- 1 January 1987
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
It is shown that non-Gaussian time series require new analysis methods to extract their structure. Spectral analysis does not seem to provide the precise information required to analyze important aspects of a time series. The conditional expected value can, in simple cases, be related to Components of the Barrett-Lampard expansion, which provides a mathematical tool for determining the system which can generate the time series.Keywords
This publication has 5 references indexed in Scilit:
- Application of a point process model to responses of cat lateral superior olive units to ipsilateral tonesHearing Research, 1986
- A new Laplace second-order autoregressive time-series model--NLAR(2)IEEE Transactions on Information Theory, 1985
- Stationary Sequences and Random FieldsPublished by Springer Nature ,1985
- Time-reversibility of linear stochastic processesJournal of Applied Probability, 1975
- An expansion for some second-order probability distributions and its application to noise problemsIEEE Transactions on Information Theory, 1955