A sequential design for determining the quantile curve in two-factor sensitivity tests

Yuxia Liu, Yubin Tian, Dianpeng Wang*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Many works in the literature have proposed methods for estimating quantiles in sensitivity experiments with a single factor. However, sensitivity experiments with multiple factors are more and more popular in practice. A significant important challenge remains the estimation of quantiles of sensitivity experiments with multiple factors, which are curves or surfaces. While some methods have been proposed to estimate the parameters in such models, the estimation of quantiles has received scant attention. Motivated by this problem and applications for field use, this study proposes a new three-phase optimal method for sensitivity experiments with two factors to estimate the interested quantiles directly. The simulation studies and the illustrative application to a pyrotechnic control subsystem demonstrate that the proposed method has excellent performance.

Original languageEnglish
Article numbere584
JournalStat
Volume12
Issue number1
DOIs
Publication statusPublished - 1 Jan 2023

Keywords

  • D-optimal
  • binary response
  • curvature
  • response surface

Fingerprint

Dive into the research topics of 'A sequential design for determining the quantile curve in two-factor sensitivity tests'. Together they form a unique fingerprint.

Cite this