Expected utility maximization and selection of stable plant cultivars

Kent M Eskridge, B. E. Johnson

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

In most plant breeding programs, selection of the best commercially suitable cultivars for a target group of environments is based on information obtained from evaluation trials cultivated in a sample of environments. Information on the performance of cultivars collected in a sample of environments can only be approximate and, consequently, selection of the best cultivar involves choosing among cultivars that respond uncertainly in many environments. The agronomic and/or economic value of a cultivar across environments may be considered the general or overall utility of the cultivar. Data from a sample of environments therefore provides only an estimate of any cultivar's overall utility, with the overall goal of selection among all cultivars being the maximization of the expected utility. Within this frame-work, expected utility maximization, an approach to decision making that has been well developed in the disciplines of economics and statistics, can assist the plant breeder in making such decisions. This research was initiated (1) to determine how expected utility maximization might be used to develop indices that are useful for selecting broadly adapted plant cultivars, and (2) to determine how the breeder's preferences might affect choice of the best cultivar. The data used in this research were from USDA Regional Soybean Tests. The results indicated that expected utility maximization, which explicitly incorporates into the selection rule the plant breeder's preferences regarding stability, can be a useful aid in the selection of stable plant cultivars.

Original languageEnglish (US)
Pages (from-to)825-832
Number of pages8
JournalTheoretical and Applied Genetics
Volume81
Issue number6
DOIs
StatePublished - Jun 1 1991

Fingerprint

cultivars
Decision Making
Economics
United States Department of Agriculture
plant breeders
Soybeans
Research
decision making
plant breeding
economic valuation
sampling
USDA
statistics
soybeans
economics
testing

Keywords

  • Genotype × environment interaction
  • Plant breeding
  • Selection
  • Utility functions
  • Yield stability

ASJC Scopus subject areas

  • Biotechnology
  • Agronomy and Crop Science
  • Genetics

Cite this

Expected utility maximization and selection of stable plant cultivars. / Eskridge, Kent M; Johnson, B. E.

In: Theoretical and Applied Genetics, Vol. 81, No. 6, 01.06.1991, p. 825-832.

Research output: Contribution to journalArticle

@article{5f242a2180f74608a25a00b558f7e960,
title = "Expected utility maximization and selection of stable plant cultivars",
abstract = "In most plant breeding programs, selection of the best commercially suitable cultivars for a target group of environments is based on information obtained from evaluation trials cultivated in a sample of environments. Information on the performance of cultivars collected in a sample of environments can only be approximate and, consequently, selection of the best cultivar involves choosing among cultivars that respond uncertainly in many environments. The agronomic and/or economic value of a cultivar across environments may be considered the general or overall utility of the cultivar. Data from a sample of environments therefore provides only an estimate of any cultivar's overall utility, with the overall goal of selection among all cultivars being the maximization of the expected utility. Within this frame-work, expected utility maximization, an approach to decision making that has been well developed in the disciplines of economics and statistics, can assist the plant breeder in making such decisions. This research was initiated (1) to determine how expected utility maximization might be used to develop indices that are useful for selecting broadly adapted plant cultivars, and (2) to determine how the breeder's preferences might affect choice of the best cultivar. The data used in this research were from USDA Regional Soybean Tests. The results indicated that expected utility maximization, which explicitly incorporates into the selection rule the plant breeder's preferences regarding stability, can be a useful aid in the selection of stable plant cultivars.",
keywords = "Genotype × environment interaction, Plant breeding, Selection, Utility functions, Yield stability",
author = "Eskridge, {Kent M} and Johnson, {B. E.}",
year = "1991",
month = "6",
day = "1",
doi = "10.1007/BF00224997",
language = "English (US)",
volume = "81",
pages = "825--832",
journal = "Theoretical And Applied Genetics",
issn = "0040-5752",
publisher = "Springer Verlag",
number = "6",

}

TY - JOUR

T1 - Expected utility maximization and selection of stable plant cultivars

AU - Eskridge, Kent M

AU - Johnson, B. E.

PY - 1991/6/1

Y1 - 1991/6/1

N2 - In most plant breeding programs, selection of the best commercially suitable cultivars for a target group of environments is based on information obtained from evaluation trials cultivated in a sample of environments. Information on the performance of cultivars collected in a sample of environments can only be approximate and, consequently, selection of the best cultivar involves choosing among cultivars that respond uncertainly in many environments. The agronomic and/or economic value of a cultivar across environments may be considered the general or overall utility of the cultivar. Data from a sample of environments therefore provides only an estimate of any cultivar's overall utility, with the overall goal of selection among all cultivars being the maximization of the expected utility. Within this frame-work, expected utility maximization, an approach to decision making that has been well developed in the disciplines of economics and statistics, can assist the plant breeder in making such decisions. This research was initiated (1) to determine how expected utility maximization might be used to develop indices that are useful for selecting broadly adapted plant cultivars, and (2) to determine how the breeder's preferences might affect choice of the best cultivar. The data used in this research were from USDA Regional Soybean Tests. The results indicated that expected utility maximization, which explicitly incorporates into the selection rule the plant breeder's preferences regarding stability, can be a useful aid in the selection of stable plant cultivars.

AB - In most plant breeding programs, selection of the best commercially suitable cultivars for a target group of environments is based on information obtained from evaluation trials cultivated in a sample of environments. Information on the performance of cultivars collected in a sample of environments can only be approximate and, consequently, selection of the best cultivar involves choosing among cultivars that respond uncertainly in many environments. The agronomic and/or economic value of a cultivar across environments may be considered the general or overall utility of the cultivar. Data from a sample of environments therefore provides only an estimate of any cultivar's overall utility, with the overall goal of selection among all cultivars being the maximization of the expected utility. Within this frame-work, expected utility maximization, an approach to decision making that has been well developed in the disciplines of economics and statistics, can assist the plant breeder in making such decisions. This research was initiated (1) to determine how expected utility maximization might be used to develop indices that are useful for selecting broadly adapted plant cultivars, and (2) to determine how the breeder's preferences might affect choice of the best cultivar. The data used in this research were from USDA Regional Soybean Tests. The results indicated that expected utility maximization, which explicitly incorporates into the selection rule the plant breeder's preferences regarding stability, can be a useful aid in the selection of stable plant cultivars.

KW - Genotype × environment interaction

KW - Plant breeding

KW - Selection

KW - Utility functions

KW - Yield stability

UR - http://www.scopus.com/inward/record.url?scp=34249923147&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34249923147&partnerID=8YFLogxK

U2 - 10.1007/BF00224997

DO - 10.1007/BF00224997

M3 - Article

VL - 81

SP - 825

EP - 832

JO - Theoretical And Applied Genetics

JF - Theoretical And Applied Genetics

SN - 0040-5752

IS - 6

ER -