Inverse regression for ridge recovery: A data-driven approach for parameter reduction in computer experiments
Preprint
Abstract
Parameter reduction can enable otherwise infeasible design and uncertainty studies with modern computational science models that contain several input parameters. In statistical regression, techniques for sufficient dimension reduction (SDR) use data to reduce the predictor dimension of a regression problem. A computational scientist hoping to use SDR for parameter reduction encounters a problem: a computer prediction is best represented by a deterministic function of the inputs, so data comprised of computer simulation queries fail to satisfy the SDR assumptions. To address this problem, we interpret SDR methods sliced inverse regression (SIR) and sliced average variance estimation (SAVE) as estimating the directions of a ridge function, which is a composition of a low-dimensional linear transformation with a nonlinear function. Within this interpretation, SIR and SAVE estimate matrices of integrals whose column spaces are contained in the ridge directions' span; we analyze and numerically verify convergence of these column spaces as the number of computer model queries increases. Moreover, we show example functions that are not ridge functions but whose inverse conditional moment matrices are low-rank. Consequently, the computational scientist should beware when using SIR and SAVE for parameter reduction, since SIR and SAVE may mistakenly suggest that truly important directions are unimportant.Keywords
All Related Versions
This publication has 23 references indexed in Scilit:
- Gauss–Christoffel quadrature for inverse regression: applications to computer experimentsStatistics and Computing, 2018
- A near-stationary subspace for ridge approximationComputer Methods in Applied Mechanics and Engineering, 2017
- Dimension reduction in magnetohydrodynamics power generation models: Dimensional analysis and active subspacesStatistical Analysis and Data Mining: An ASA Data Science Journal, 2017
- Efficient evaluation of small failure probability in high‐dimensional groundwater contaminant transport modeling via a two‐stage Monte Carlo methodWater Resources Research, 2017
- Inverse regression-based uncertainty quantification algorithms for high-dimensional models: Theory and practiceJournal of Computational Physics, 2016
- Learning non-parametric basis independent models from point queries via low-rank methodsApplied and Computational Harmonic Analysis, 2014
- Learning Functions of Few Arbitrary Linear Parameters in High DimensionsFoundations of Computational Mathematics, 2012
- Capturing Ridge Functions in High Dimensions from Point QueriesConstructive Approximation, 2011
- Successive direction extraction for estimating the central subspace in a multiple-index regressionJournal of Multivariate Analysis, 2008
- A Taxonomy of Global Optimization Methods Based on Response SurfacesJournal of Global Optimization, 2001