This article is concerned with the study of statistical power in agent-based modeling (ABM). After an overview of classic statistics theory on how to interpret Type-II error (whose occurrence is also referred to as a false negative) and power, the manuscript presents a study on ABM simulation articles published in management journals and other outlets likely to publish management and organizational research. Findings show that most studies are underpowered, with some being overpowered. After discussing the risks of under- and overpower, we present two formulas to approximate the number of simulation runs to reach an appropriate level of power. The study concludes with the importance for organizational behavior scholars to perform their models in an attempt to reach a power of 0.95 or higher at the 0.01 significance level.
This article proposes a framework for the analysis of experienced discrimination in home mortgages. It addresses the problem of home mortgage lending discrimination in one of the richest areas of northern Italy. Employees of a local hospital were interviewed to study their perception (or experience) of discriminatory behavior related to home financing. The analysis follows two steps. The first evaluates self-selection (the probability that individuals apply) and the second focuses on the likelihood that applications are accepted by the bank. Findings show that discrimination is likely to appear when the applicant's nationality is considered. In addition to its findings, the study (a) provides an original econometric model on a two-step procedure to test perceived discrimination and (b) suggests a method and approach that may constitute a point of reference for those willing to study perceived discrimination.
This chapter is an attempt to answer the question "how many runs of a computational simulation should one do," and it gives an answer by means of statistical analysis. After defining the nature of the problem and which types of simulation are mostly affected by it, the chapter introduces statistical power analysis as a way to determine the appropriate number of runs. Two examples are then produced using results from an agent-based model. The reader is then guided through the application of this statistical technique and exposed to its limits and potentials.
The aim of this article is to present an approach to the analysis of simple systems composed of a large number of units in interaction. Suppose to have a large number of agents belonging to a finite number of different groups: as the agents randomly interact with each other, they move from a group to another as a result of the interaction. The object of interest is the stochastic process describing the number of agents in each group. As this is generally intractable, it has been proposed in the literature to approximate it in several ways. We review these approximations and we illustrate them with reference to a version of the epidemic model. The tools presented in the paper should be considered as a complement rather than as a substitute of the classical analysis of ABMs through simulation.