Sample size estimation for power and accuracy in the experimental comparison of algorithms

Abstract

Experimental comparisons of performance represent an important aspect of research on optimization algorithms. In this work we present a methodology for defining the required sample sizes for designing experiments with desired statistical properties for the comparison of two methods on a given problem class. The proposed approach allows the experimenter to define desired levels of accuracy for estimates of mean performance differences on individual problem instances, as well as the desired statistical power for comparing mean performances over a problem class of interest. The method calculates the required number of problem instances, and runs the algorithms on each test instance so that the accuracy of the estimated differences in performance is controlled at the predefined level. Two examples illustrate the application of the proposed method, and its ability to achieve the desired statistical properties with a methodologically sound definition of the relevant sample sizes.

Publication DOI: https://doi.org/10.1007/s10732-018-9396-7
Additional Information: © Springer Nature B.V. 2018. The final publication is available at Springer via http://dx.doi.org/10.1007/s10732-018-9396-7
Uncontrolled Keywords: Accuracy of parameter estimation,Experimental comparison of algorithms,Iterative sampling,Sample size estimation,Statistical methods,Software,Information Systems,Computer Networks and Communications,Control and Optimization,Management Science and Operations Research,Artificial Intelligence
Publication ISSN: 1381-1231
Last Modified: 29 Apr 2024 07:29
Date Deposited: 14 Mar 2019 14:50
Full Text Link: https://arxiv.o ... /pdf/1808.02997
Related URLs: https://link.sp ... 0732-018-9396-7 (Publisher URL)
http://www.scop ... tnerID=8YFLogxK (Scopus URL)
PURE Output Type: Article
Published Date: 2019-04-15
Published Online Date: 2018-10-04
Accepted Date: 2018-09-26
Authors: Campelo, Felipe (ORCID Profile 0000-0001-8432-4325)
Takahashi, Fernanda

Download

[img]

Version: Accepted Version

| Preview

Export / Share Citation


Statistics

Additional statistics for this record