Abstract
We develop algorithms for sampling from a probability distribution on a submanifold embedded in $\mathbb{R}^{n}$. Applications are given to the evaluation of algorithms in ‘Topological Statistics’; to goodness of fit tests in exponential families and to Neyman’s smooth test. This article is partially expository, giving an introduction to the tools of geometric measure theory.
Information
Published: 1 January 2013
First available in Project Euclid: 23 September 2013
zbMATH: 1356.62015
Digital Object Identifier: 10.1214/12-IMSCOLL1006
Subjects:
Primary:
60K35
,
60K35
Secondary:
60K35
Keywords:
conditional distribution
,
geometric measure theory
,
Manifold
,
sampling
Rights: Copyright © 2013, Institute of Mathematical Statistics
