Bandwidth Selection for Multivariate Kernel Density Estimation Using MCMC

Xibin Zhang (), Maxwell L. King () and Rob J Hyndman ()

No 9/04, Monash Econometrics and Business Statistics Working Papers from Monash University, Department of Econometrics and Business Statistics

Abstract: We provide Markov chain Monte Carlo (MCMC) algorithms for computing the bandwidth matrix for multivariate kernel density estimation. Our approach is based on treating the elements of the bandwidth matrix as parameters to be estimated, which we do by optimizing the likelihood cross-validation criterion. Numerical results show that the resulting bandwidths are superior to all existing methods; for dimensions greater than two, our algorithm is the first practical method for estimating the optimal bandwidth matrix. Moreover, the MCMC algorithm for bandwidth selection for multivariate data has no increased difficulty as the dimension of data increases.

Keywords: Bandwidth selection; cross-validation; multivariate kernel density estimation; sampling algorithms. (search for similar items in EconPapers)
JEL-codes: C14 C11 C51 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-ecm and nep-ets
Date: 2004-04
View list of references View citations in EconPapers

Downloads: (external link)
http://www.buseco.monash.edu.au/depts/ebs/pubs/wpapers/2004/wp9-04.pdf (application/pdf)

Related works:
Working Paper: Bandwidth Selection for Multivariate Kernel Density Estimation Using MCMC (2004)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: http://EconPapers.repec.org/RePEc:msh:ebswps:2004-9

Ordering information: This working paper can be ordered from
http://www.buseco.mo ... ts/ebs/pubs/wpapers/

Access Statistics for this paper

More papers in Monash Econometrics and Business Statistics Working Papers from Monash University, Department of Econometrics and Business Statistics
Address: PO Box 11E, Monash University, Victoria 3800, Australia
Contact information at EDIRC.
Series data maintained by Simone Grose ().