The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here: https://www.microsoft.com/en-us/microsoft-365/windows/end-of-ie-support).

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Aspects of multilinear algebra in statistical analysis of seasonal multivariate time series

På svenska

Author

Summary, in English

Representations of concepts from multivariate statistics are studied using multilinear algebra. Applications are given to the analysis of seasonal multivariate time series.



The first paper concerns a class of permutation matrices which are representations of permutation operators on tensor spaces.



In the second paper, the derivation of moments and cumulants of Hilbert space valued random variables are studied. Relations between moments and cumulants - as vector space elements - are given which generalize well-known results for scalar random variables. Representations of differentials are discussed and results for differentials of matrix valued functions with matrix arguments are derived. Moments and cumulants of the non-central Wishart distribution are given.



in the third paper, seasonal multivariate time series are investigated using the concepts described in the first two papers. Estimators of unknown parameters are given and their asymptotic distributions are derived. Test statistics for testing homogeneity in time and independence of subprocesses are given and also their asymptotic distributions.

Publishing year

1985-06-07

Language

English

Document type

Dissertation

Publisher

Department of Mathematical Statistics, Lund University

Topic

  • Probability Theory and Statistics

Status

Published

Supervisor

  • Gunnar Blom

Defence date

7 June 1985

Defence time

10:00

Defence place

MH:C

Opponent

  • Morris L Eaton (Professor)