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

Download:

Current browse context:

math.ST

Change to browse by:

References & Citations

Bookmark

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

Mathematics > Statistics Theory

Title: Derivative-based global sensitivity measures and their link with Sobol sensitivity indices

Abstract: The variance-based method of Sobol sensitivity indices is very popular among practitioners due to its efficiency and easiness of interpretation. However, for high-dimensional models the direct application of this method can be very time consuming and prohibitively expensive to use. One of the alternative global sensitivity analysis methods known as the method of derivative based global sensitivity measures (DGSM) has recently become popular among practitioners. It has a link with the Morris screening method and Sobol sensitivity indices. DGSM are very easy to implement and evaluate numerically. The computational time required for numerical evaluation of DGSM is generally much lower than that for estimation of Sobol sensitivity indices. We present a survey of recent advances in DGSM and new results concerning new lower and upper bounds on the values of Sobol total sensitivity indices. Using these bounds it is possible in most cases to get a good practical estimation of the values of Sobol total sensitivity indices. Several examples are used to illustrate an application of DGSM.
Comments: Monte Carlo and Quasi-Monte Carlo Methods 2014, conference, Monte Carlo and Quasi-Monte Carlo Methods, R. Cools and D. Nuyens (eds.), Springer Proceedings in Mathematics & Statistics 163, Springer International Publishing Switzerland 2016
Subjects: Statistics Theory (math.ST); Numerical Analysis (math.NA); Probability (math.PR)
DOI: 10.1007/978-3-319-33507-0_23
Cite as: arXiv:1605.07830 [math.ST]
  (or arXiv:1605.07830v1 [math.ST] for this version)

Submission history

From: Sergei Kucherenko [view email]
[v1] Wed, 25 May 2016 11:17:22 GMT (192kb)

Link back to: arXiv, form interface, contact.