We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.CO

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Computation

Title: Stochastic optimization for numerical evaluation of imprecise probabilities

Abstract: In applications of imprecise probability, analysts must compute lower (or upper) expectations, defined as the infimum of an expectation over a set of parameter values. Monte Carlo methods consistently approximate expectations at fixed parameter values, but can be costly to implement in grid search to locate minima over large subsets of the parameter space. We investigate the use of stochastic iterative root-finding methods for efficiently computing lower expectations. In two examples we illustrate the use of various stochastic approximation methods, and demonstrate their superior performance in comparison to grid search.
Comments: 18 pages, 6 figures, 1 table
Subjects: Computation (stat.CO)
MSC classes: 62-08
Cite as: arXiv:2103.02659 [stat.CO]
  (or arXiv:2103.02659v1 [stat.CO] for this version)

Submission history

From: Nicholas Syring [view email]
[v1] Wed, 3 Mar 2021 20:15:18 GMT (944kb)

Link back to: arXiv, form interface, contact.