International Journal of System Dynamics Applications, 2(1), 97-113, January-March 2013 97
Copyright © 2013, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Keywords: Discrete Probability, Probability Proportional to Size Sampling, Random Sampling, Stochastic
Optimization
1. INTRODUCTION
Many problems that use simulation methods need to generate random samples with certain
probability distributions. There are several proposed methods that generate random samples
with a defined probability distribution. These methods need to generate the most representative
samples of the required probability distribution; this can be a challenge when the sample size
needs to be small. For example, in stochastic optimization problems, minimizing the number
of samples is very important. Note that the more samples, the more time is needed to evaluate
A Novel Quota Sampling
Algorithm for Generating
Representative Random Samples
given Small Sample Size
Ahmed M. Fouad, Department of Computer Engineering, Cairo University, Cairo, Giza,
Egypt
Mohamed Saleh, Department of Operations Research and Decision Support, Cairo University,
Cairo, Giza, Egypt
Amir F.Atiya, Department of Computer Engineering, Cairo University, Cairo, Giza, Egypt
ABSTRACT
In this paper, a novel algorithm is proposed for sampling from discrete probability distributions using the
probability proportional to size sampling method, which is a special case of Quota sampling method. The
motivation for this study is to devise an eficient sampling algorithm that can be used in stochastic optimiza-
tion problems -- when there is a need to minimize the sample size. Several experiments have been conducted
to compare the proposed algorithm with two widely used sample generation methods, the Monte Carlo using
inverse transform, and quasi-Monte Carlo algorithms. The proposed algorithm gave better accuracy than
these methods, and in terms of time complexity it is nearly of the same order.
DOI: 10.4018/ijsda.2013010105