Bootstrapping and Conditional Simulation in Kriging: Better Confidence Intervals and Optimization (Replaced by CentER DP 2014-076)

Research output: Working paperDiscussion paperOther research output

37 Downloads (Pure)

Abstract

Abstract: This paper investigates two related questions: (1) How to derive a confidence interval for the output of a combination of simulation inputs not yet simulated? (2) How to select the next combination to be simulated when searching for the optimal combination? To answer these questions, the paper uses parametric bootstrapped Kriging and "conditional simulation". Classic Kriging estimates the variance of its predictor by plugging-in the estimated GP parameters so this variance is biased. The main conclusion is that classic Kriging seems quite robust; i.e., classic Kriging gives acceptable confidence intervals and estimates of the optimal solution.
Original languageEnglish
Place of PublicationTilburg
PublisherInformation Management
Number of pages24
Volume2013-038
Publication statusPublished - 2013

Publication series

NameCentER Discussion Paper
Volume2013-038

Fingerprint

Confidence interval
Simulation
Kriging
Bootstrapping
Optimal solution
Predictors

Keywords

  • Simulation
  • Optimization
  • Kriging
  • Bootstrap

Cite this

@techreport{9c95782eca4c44a5893a38edea453b6b,
title = "Bootstrapping and Conditional Simulation in Kriging: Better Confidence Intervals and Optimization (Replaced by CentER DP 2014-076)",
abstract = "Abstract: This paper investigates two related questions: (1) How to derive a confidence interval for the output of a combination of simulation inputs not yet simulated? (2) How to select the next combination to be simulated when searching for the optimal combination? To answer these questions, the paper uses parametric bootstrapped Kriging and {"}conditional simulation{"}. Classic Kriging estimates the variance of its predictor by plugging-in the estimated GP parameters so this variance is biased. The main conclusion is that classic Kriging seems quite robust; i.e., classic Kriging gives acceptable confidence intervals and estimates of the optimal solution.",
keywords = "Simulation, Optimization, Kriging, Bootstrap",
author = "E. Mehdad and Kleijnen, {Jack P.C.}",
note = "Pagination: 24",
year = "2013",
language = "English",
volume = "2013-038",
series = "CentER Discussion Paper",
publisher = "Information Management",
type = "WorkingPaper",
institution = "Information Management",

}

Bootstrapping and Conditional Simulation in Kriging : Better Confidence Intervals and Optimization (Replaced by CentER DP 2014-076). / Mehdad, E.; Kleijnen, Jack P.C.

Tilburg : Information Management, 2013. (CentER Discussion Paper; Vol. 2013-038).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Bootstrapping and Conditional Simulation in Kriging

T2 - Better Confidence Intervals and Optimization (Replaced by CentER DP 2014-076)

AU - Mehdad, E.

AU - Kleijnen, Jack P.C.

N1 - Pagination: 24

PY - 2013

Y1 - 2013

N2 - Abstract: This paper investigates two related questions: (1) How to derive a confidence interval for the output of a combination of simulation inputs not yet simulated? (2) How to select the next combination to be simulated when searching for the optimal combination? To answer these questions, the paper uses parametric bootstrapped Kriging and "conditional simulation". Classic Kriging estimates the variance of its predictor by plugging-in the estimated GP parameters so this variance is biased. The main conclusion is that classic Kriging seems quite robust; i.e., classic Kriging gives acceptable confidence intervals and estimates of the optimal solution.

AB - Abstract: This paper investigates two related questions: (1) How to derive a confidence interval for the output of a combination of simulation inputs not yet simulated? (2) How to select the next combination to be simulated when searching for the optimal combination? To answer these questions, the paper uses parametric bootstrapped Kriging and "conditional simulation". Classic Kriging estimates the variance of its predictor by plugging-in the estimated GP parameters so this variance is biased. The main conclusion is that classic Kriging seems quite robust; i.e., classic Kriging gives acceptable confidence intervals and estimates of the optimal solution.

KW - Simulation

KW - Optimization

KW - Kriging

KW - Bootstrap

M3 - Discussion paper

VL - 2013-038

T3 - CentER Discussion Paper

BT - Bootstrapping and Conditional Simulation in Kriging

PB - Information Management

CY - Tilburg

ER -