Mixtures of triangular densities with applications to Bayesian mode regressions

Simple item page
dc.contributor.advisorDamien, Paul, 1960-
dc.creatorHo, Chi-Sanen
dc.date.accessioned2014-09-22T17:51:21Zen
dc.date.accessioned2018-01-22T22:26:34Z
dc.date.available2018-01-22T22:26:34Z
dc.date.issued2014-08en
dc.date.submittedAugust 2014en
dc.date.updated2014-09-22T17:51:21Zen
dc.descriptiontexten
dc.description.abstractThe main focus of this thesis is to develop full parametric and semiparametric Bayesian inference for data arising from triangular distributions. A natural consequence of working with such distributions is it allows one to consider regression models where the response variable is now the mode of the data distribution. A new family of nonparametric prior distributions is developed for a certain class of convex densities of particular relevance to mode regressions. Triangular distributions arise in several contexts such as geosciences, econometrics, finance, health care management, sociology, reliability engineering, decision and risk analysis, etc. In many fields, experts, typically, have a reasonable idea about the range and most likely values that define a data distribution. Eliciting these quantities is thus, generally, easier than eliciting moments of other commonly known distributions. Using simulated and actual data, applications of triangular distributions, with and without mode regressions, in some of the aforementioned areas are tackled.en
dc.description.departmentInformation, Risk, and Operations Management (IROM)en
dc.format.mimetypeapplication/pdfen
dc.identifier.urihttp://hdl.handle.net/2152/26077en
dc.language.isofren
dc.subjectBayesian inferenceen
dc.subjectConsistencyen
dc.subjectConvex densitiesen
dc.subjectTriangular distributionsen
dc.titleMixtures of triangular densities with applications to Bayesian mode regressionsen
dc.typeThesisen

Files

Collections

University of Texas at Austin