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

Download:

Current browse context:

stat.ME

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 > Methodology

Title: Assessing variable activity for Bayesian regression trees

Authors: Akira Horiguchi (1), Matthew T. Pratola (1), Thomas J. Santner (1) ((1) Department of Statistics, The Ohio State University)
Abstract: Bayesian Additive Regression Trees (BART) are non-parametric models that can capture complex exogenous variable effects. In any regression problem, it is often of interest to learn which variables are most active. Variable activity in BART is usually measured by counting the number of times a tree splits for each variable. Such one-way counts have the advantage of fast computations. Despite their convenience, one-way counts have several issues. They are statistically unjustified, cannot distinguish between main effects and interaction effects, and become inflated when measuring interaction effects. An alternative method well-established in the literature is Sobol' indices, a variance-based global sensitivity analysis technique. However, these indices often require Monte Carlo integration, which can be computationally expensive. This paper provides analytic expressions for Sobol' indices for BART posterior samples. These expressions are easy to interpret and are computationally feasible. Furthermore, we will show a fascinating connection between first-order (main-effects) Sobol' indices and one-way counts. We also introduce a novel ranking method, and use this to demonstrate that the proposed indices preserve the Sobol'-based rank order of variable importance. Finally, we compare these methods using analytic test functions and the En-ROADS climate impacts simulator.
Comments: 46 pages, 8 figures, submitted to the special issue "Recent Advances in Sensitivity Analysis of Model Outputs" in the Reliability Engineering and Safety System journal
Subjects: Methodology (stat.ME)
Cite as: arXiv:2005.13622 [stat.ME]
  (or arXiv:2005.13622v2 [stat.ME] for this version)

Submission history

From: Akira Horiguchi [view email]
[v1] Wed, 27 May 2020 20:12:10 GMT (167kb,D)
[v2] Tue, 15 Sep 2020 02:47:05 GMT (159kb,D)

Link back to: arXiv, form interface, contact.