Uncertainty Quantification and Global Sensitivity Analysis for Economic Models

49 Pages Posted: 25 Jan 2017 Last revised: 29 Dec 2017

See all articles by Daniel Harenberg

Daniel Harenberg

ETH Zürich - CER-ETH - Center of Economic Research at ETH Zurich; Oxford Economics

Stefano Marelli

ETH Zurich - Chair of Risk, Safety and Uncertainty Quantification

Bruno Sudret

ETH Zurich

Viktor Winschel

ETH Zürich

Multiple version iconThere are 2 versions of this paper

Date Written: December 29, 2017

Abstract

We present a global sensitivity analysis that quantifies the impact of parameter uncertainty on model outcomes. Specifically, we propose variance-decomposition-based Sobol' indices to establish an importance ranking of parameters and univariate effects to determine the direction of their impact. We employ the state-of-the-art approach of constructing a polynomial chaos expansion of the model, from which Sobol' indices and univariate effects are then obtained analytically, using only a limited number of model evaluations. We apply this analysis to several quantities of interest of a standard real-business-cycle model and compare it to traditional local sensitivity analysis approaches. The results show that local sensitivity analysis can be very misleading, whereas the proposed method accurately and efficiently ranks all parameters according to importance, identifying interactions and nonlinearities.

Keywords: computational techniques, uncertainty quantification, sensitivity analysis, polynomial chaos expansion

JEL Classification: C60, C63

Suggested Citation

Harenberg, Daniel and Marelli, Stefano and Sudret, Bruno and Winschel, Viktor, Uncertainty Quantification and Global Sensitivity Analysis for Economic Models (December 29, 2017). Available at SSRN: https://ssrn.com/abstract=2903994 or http://dx.doi.org/10.2139/ssrn.2903994

Daniel Harenberg (Contact Author)

ETH Zürich - CER-ETH - Center of Economic Research at ETH Zurich ( email )

Zürichbergstrasse 18
Zurich, 8092
Switzerland

Oxford Economics ( email )

United Kingdom

Stefano Marelli

ETH Zurich - Chair of Risk, Safety and Uncertainty Quantification ( email )

Stefano-Franscini-Platz 5
Zurich, Zurich 8093
Switzerland
+416330670 (Phone)

HOME PAGE: http://www.rsuq.ethz.ch

Bruno Sudret

ETH Zurich ( email )

Rämistrasse 101
ZUE F7
Zürich, 8092
Switzerland

Viktor Winschel

ETH Zürich ( email )

Zürichbergstrasse 18
Zürich, Zürich 8'32
Switzerland
+49-1797621055 (Phone)

Register to save articles to
your library

Register

Paper statistics

Downloads
144
Abstract Views
667
rank
160,803
PlumX Metrics