Title

Novel Methods for Maximizing and Evaluating Adaptive Measurement Efficiency

Investigators

Alan B. Cobo-Lewis, Principal Investigator; University of Maine, OronoFollow

Additional Participants

Graduate Student

Liying Tan

Research Experience for Undergraduates

Tasha Smallwood
Lara Than
John Montani

Project Period

July 1, 1998-May 31, 2002

Level of Access

Open-Access Report

Grant Number

9896277

Submission Date

2-12-2004

Abstract

Contributions within Discipline: The findings have improved the efficiency of adaptive measurement in psychophysics, in experimental paradigms where individual trials are often information-poor and experiments are consequently long. The Bayesian adaptive methodology improves the information throughput in such experiments and improves on heuristic methods. The multivariate estimation also extends the utility of Bayesian adaptive estimation into realms where it is even more important because of the 'curse of dimensionality' (where the size of parameter space is exponential in the number of parameters). In addition, the work on nonparametric adaptive methods has helped reveal the source of bias in simpler adaptive methodology that has often incorrectly been taken to be safe because of its apparent lack of statistical assumptions. By revealint the source of such bias, it offers solutions for minimizing the bias.

Rights and Access Note

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. In addition, no permission is required from the rights-holder(s) for educational uses. For other uses, you need to obtain permission from the rights-holder(s).

Recommended Citation

Cobo-Lewis, Alan B., "Novel Methods for Maximizing and Evaluating Adaptive Measurement Efficiency" (2004). University of Maine Office of Research Administration: Grant Reports. 107.
https://digitalcommons.library.umaine.edu/orsp_reports/107