Computational Methods for Measuring the Difference of Empirical Distributions

Posted: 15 Dec 2005

See all articles by Gregory L. Poe

Gregory L. Poe

Cornell University - School of Applied Economics and Management; PERC - Property and Environment Research Center

Kelly Giraud

University of New Hampshire - Department of Resource Economics and Development

John B. Loomis

Colorado State University, Fort Collins - Department of Agriculture and Resource Economics

Abstract

This paper presents a simple computational method for measuring the difference of independent empirical distributions estimated by bootstrapping or other resampling approaches. Using data from a field test of external scope in contingent valuation, this complete combinatorial method is compared with other methods (empirical convolutions, repeated sampling, normality, nonoverlapping confidence intervals) that have been suggested in the literature. Tradeoffs between methods are discussed in terms of programming complexity, time and computer resources required, bias, and the precision of the estimate.

Suggested Citation

Poe, Gregory L. and Giraud, Kelly and Loomis, John B., Computational Methods for Measuring the Difference of Empirical Distributions. American Journal of Agricultural Economics, Vol. 87, No. 2, pp. 353-365, May 2005, Available at SSRN: https://ssrn.com/abstract=856497 or http://dx.doi.org/10.1111/j.1467-8276.2005.00727.x

Gregory L. Poe (Contact Author)

Cornell University - School of Applied Economics and Management ( email )

248 Warren Hall
Ithaca, NY 14853
United States
607-255-4707 (Phone)
607-255-9984 (Fax)

PERC - Property and Environment Research Center

2048 Analysis Drive
Suite A
Bozeman, MT 59718
United States

Kelly Giraud

University of New Hampshire - Department of Resource Economics and Development ( email )

15 College Road
Durham, NH 03824
United States

John B. Loomis

Colorado State University, Fort Collins - Department of Agriculture and Resource Economics ( email )

B310 Andrew G. Clark
Fort Collins, CO 80523
United States
970-491-2485 (Phone)

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
768
PlumX Metrics