A genetic algorithm approach to optimal stratified sampling

Patrick R. McMullen, David Albritton

Research output: Contribution to conferencePaper

Abstract

A technique is presented to assist in the design of stratified random sampling plans via a genetic algorithm approach. When limited resources are available, researchers must often be thrifty in their attempt to find a minimized variance estimate of the mean, subject to the cost constraints of collecting data. The research presented here exploits the artificial intelligence technique of genetic algorithms to find stratified sampling plans, and preliminary results are near-optimal.

Original languageEnglish (US)
Pages1663-1668
Number of pages6
StatePublished - Dec 1 2003
Externally publishedYes
Event34th Annual Meeting of the Decision Sciences Institute - Washington, DC, United States
Duration: Nov 22 2003Nov 25 2003

Other

Other34th Annual Meeting of the Decision Sciences Institute
CountryUnited States
CityWashington, DC
Period11/22/0311/25/03

Fingerprint

Genetic algorithms
Sampling
Artificial intelligence
Costs
Genetic algorithm
Stratified sampling
Resources
Random sampling

Keywords

  • Heuristics
  • Mathematical Programming/Optimization
  • Simulation

ASJC Scopus subject areas

  • Hardware and Architecture
  • Management Information Systems

Cite this

McMullen, P. R., & Albritton, D. (2003). A genetic algorithm approach to optimal stratified sampling. 1663-1668. Paper presented at 34th Annual Meeting of the Decision Sciences Institute, Washington, DC, United States.

A genetic algorithm approach to optimal stratified sampling. / McMullen, Patrick R.; Albritton, David.

2003. 1663-1668 Paper presented at 34th Annual Meeting of the Decision Sciences Institute, Washington, DC, United States.

Research output: Contribution to conferencePaper

McMullen, PR & Albritton, D 2003, 'A genetic algorithm approach to optimal stratified sampling' Paper presented at 34th Annual Meeting of the Decision Sciences Institute, Washington, DC, United States, 11/22/03 - 11/25/03, pp. 1663-1668.
McMullen PR, Albritton D. A genetic algorithm approach to optimal stratified sampling. 2003. Paper presented at 34th Annual Meeting of the Decision Sciences Institute, Washington, DC, United States.
McMullen, Patrick R. ; Albritton, David. / A genetic algorithm approach to optimal stratified sampling. Paper presented at 34th Annual Meeting of the Decision Sciences Institute, Washington, DC, United States.6 p.
@conference{177a114552f84034a500e7ef4bdc6f48,
title = "A genetic algorithm approach to optimal stratified sampling",
abstract = "A technique is presented to assist in the design of stratified random sampling plans via a genetic algorithm approach. When limited resources are available, researchers must often be thrifty in their attempt to find a minimized variance estimate of the mean, subject to the cost constraints of collecting data. The research presented here exploits the artificial intelligence technique of genetic algorithms to find stratified sampling plans, and preliminary results are near-optimal.",
keywords = "Heuristics, Mathematical Programming/Optimization, Simulation",
author = "McMullen, {Patrick R.} and David Albritton",
year = "2003",
month = "12",
day = "1",
language = "English (US)",
pages = "1663--1668",
note = "34th Annual Meeting of the Decision Sciences Institute ; Conference date: 22-11-2003 Through 25-11-2003",

}

TY - CONF

T1 - A genetic algorithm approach to optimal stratified sampling

AU - McMullen, Patrick R.

AU - Albritton, David

PY - 2003/12/1

Y1 - 2003/12/1

N2 - A technique is presented to assist in the design of stratified random sampling plans via a genetic algorithm approach. When limited resources are available, researchers must often be thrifty in their attempt to find a minimized variance estimate of the mean, subject to the cost constraints of collecting data. The research presented here exploits the artificial intelligence technique of genetic algorithms to find stratified sampling plans, and preliminary results are near-optimal.

AB - A technique is presented to assist in the design of stratified random sampling plans via a genetic algorithm approach. When limited resources are available, researchers must often be thrifty in their attempt to find a minimized variance estimate of the mean, subject to the cost constraints of collecting data. The research presented here exploits the artificial intelligence technique of genetic algorithms to find stratified sampling plans, and preliminary results are near-optimal.

KW - Heuristics

KW - Mathematical Programming/Optimization

KW - Simulation

UR - http://www.scopus.com/inward/record.url?scp=0442325484&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0442325484&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0442325484

SP - 1663

EP - 1668

ER -