Abstracts - faqs.org

Abstracts

Business, general

Search abstracts:
Abstracts » Business, general

Model selection criteria: an investigation of relative accuracy, posterior probabilities, and combinations of criteria

Article Abstract:

The use of empirical criteria in the evaluation and selection of quantitative models from a candidate set is investigated. This is done through a parameter-based simulation that is used to determine which criteria are most reliable in identifying the right model specification. This simulation, which is made up of both nonnested and nested linear regression models, is also used to obtain probability estimates of the alternative models in the candidate set. It is then used to assess the relative information content, bias and accuracy of the probabilities derived. A method for securing a joint prediction from various criteria combinations is also proposed. The method's effect on the improvement of selection reliability is evaluated as well. Based on simulation results, it is determined that leading criteria perform well in best model selection and that several of these criteria are useful in providing accurate probabilities of model superiority.

Author: Rust, Roland T., Brodie, Roderick J., Simester, Duncan, Nilikant, V.
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1995
Research, Bayesian statistical decision theory, Bayesian analysis, Mathematical models

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Eliciting von Neumann-Morgenstern utilities when probabilities are distorted or unknown

Article Abstract:

The tradeoff method is proposed as a tool for eliciting von Neumann-Morgenstern utilities in decision under risk or uncertainty. Otherwise known as the gamble-tradeoff method, it is robust against probability distortions and misconceptions, which usually result in violations of expected utility and produce inconsistencies in utility elicitation. The method therefore maintains complete validity under prospect theory, rank-dependent utility, and the cumulative prospect theory. When tested for monetary outcomes and for outcomes describing life-duration, the method shows higher risk aversion for life duration although it engenders similar utility curvature. The higher risk aversion for life duration seems to be the result of greater deviations from expected utility.

Author: Wakker, Peter, Deneffe, Daniel
Publisher: Institute for Operations Research and the Management Sciences
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1996
Analysis, Utility theory, Utility functions, Von Neumann algebras

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Similar abstracts:
  • Abstracts: An empirical investigation of the linkages among relationalism, environmental uncertainty, and bureaucratization
  • Abstracts: An empirical investigation into factors relating to the adoption of executive information systems: an analysis of EIS for collaboration and decision support
  • Abstracts: Managers as monitors: An analysis of the non-executive role of senior executives in UK companies. Perceived environmental uncertainty, entry mode choice and satisfaction with EC-MNC performance
  • Abstracts: A microeconomic model of the focused factory-vertical integration strategic decision problem. Decision making in a standby service system
  • Abstracts: Models of innovation: startups and mature corporations. Managing formation process in R&D consortia. Relational Quality: Managing Trust in Corporate Alliances
This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.
Some parts © 2025 Advameg, Inc.