Please use this identifier to cite or link to this item:
|Title:||Parameter expansion for estimation of reduced rank covariance matrices||Contributor(s):||Meyer, Karin (author)||Publication Date:||2008||DOI:||10.1186/1297-9686-40-1-3||Handle Link:||https://hdl.handle.net/1959.11/3075||Abstract:||Parameter expanded and standard expectation maximisation algorithms are described for reduced rank estimation of covariance matrices by restricted maximum likelihood, fitting the leading principal components only. Convergence behaviour of these algorithms is examined for several examples and contrasted to that of the average information algorithm, and implications for practical analyses are discussed. It is shown that expectation maximisation type algorithms are readily adapted to reduced rank estimation and converge reliably. However, as is well known for the full rank case, the convergence is linear and thus slow. Hence, these algorithms are most useful in combination with the quadratically convergent average information algorithm, in particular in the initial stages of an iterative solution scheme.||Publication Type:||Journal Article||Source of Publication:||Genetics Selection Evolution, 40(1), p. 3-24||Publisher:||BioMed Central Ltd||Place of Publication:||London, UK||ISSN:||0999-193X||Field of Research (FOR):||060412 Quantitative Genetics (incl Disease and Trait Mapping Genetics)||Peer Reviewed:||Yes||HERDC Category Description:||C1 Refereed Article in a Scholarly Journal||Statistics to Oct 2018:||Visitors: 90
|Appears in Collections:||Journal Article|
Files in This Item:
checked on Feb 8, 2019
Items in Research UNE are protected by copyright, with all rights reserved, unless otherwise indicated.