Posterior Consistency in Conditional Density Estimation by Covariate Dependent Mixtures

Andriy Norets and Justinas Pelenis

No 282, Economics Series from Institute for Advanced Studies

Abstract: This paper considers Bayesian nonparametric estimation of conditional densities by countable mixtures of location-scale densities with covariate dependent mixing probabilities. The mixing probabilities are modeled in two ways. First, we consider finite covariate dependent mixture models, in which the mixing probabilities are proportional to a product of a constant and a kernel and a prior on the number of mixture components is specified. Second, we consider kernel stick-breaking processes for modeling the mixing probabilities. We show that the posterior in these two models is weakly and strongly consistent for a large class of data generating processes.

Keywords: Bayesian nonparametrics; posterior consistency; conditional density estimation; mixtures of normal distributions; location-scale mixtures; smoothly mixing regressions; mixtures of experts; dependent Dirichlet process; kernel stick-breaking process (search for similar items in EconPapers)
JEL-codes: C11 C14 (search for similar items in EconPapers)
Pages: 26 pages
Date: 2011-12
New Economics Papers: this item is included in nep-ecm
References: Add references at CitEc
Citations: View citations in EconPapers (4)

Downloads: (external link)
https://irihs.ihs.ac.at/id/eprint/2108 First version, 2011 (application/pdf)

Related works:
Journal Article: POSTERIOR CONSISTENCY IN CONDITIONAL DENSITY ESTIMATION BY COVARIATE DEPENDENT MIXTURES (2014)
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: https://EconPapers.repec.org/RePEc:ihs:ihsesp:282

Ordering information: This working paper can be ordered from
Institute for Advanced Studies - Library, Josefstädterstr. 39, A-1080 Vienna, Austria

Access Statistics for this paper

More papers in Economics Series from Institute for Advanced Studies Josefstädterstr. 39, A-1080 Vienna, Austria. Contact information at EDIRC.
Bibliographic data for series maintained by Doris Szoncsitz ().