Discrete Approximation of a Mixture Distribution via Restricted Divergence
2017 | journal article. A publication with affiliation to the University of Göttingen.
Jump to: Cite & Linked | Documents & Media | Details | Version history
Documents & Media
article876.38 kBAdobe PDFpreprint version185.99 kBAdobe PDFR-code for example 5.14.62 kBUnknownR-code for example 5.24.61 kBUnknownR-code for example 5.26.56 kBUnknown
Details
- Authors
- Roever, Christian ; Friede, Tim
- Abstract
- Mixture distributions arise inmany application areas, for example, as marginal distributions or convolutions of distributions. We present amethod of constructing an easily tractable discretemixture distribution as an approximation to a mixture distribution with a large to infinite number, discrete or continuous, of components. The proposed DIRECT (divergence restricting conditional tesselation) algorithm is set up such that a prespecified precision, defined in terms of Kullback-Leibler divergence between true distribution and approximation, is guaranteed. Application of the algorithm is demonstrated in two examples. Supplementary materials for this article are available online.
- Issue Date
- 2017
- Status
- published
- Publisher
- Amer Statistical Assoc
- Journal
- Journal of Computational and Graphical Statistics
- ISSN
- 1537-2715; 1061-8600
- Sponsor
- European Union [FP HEALTH 2013-602144]