Record Details


Field	Value

Title	Replication data for: A Statistical Method for Empirical Testing of Competing Theories

Identifier	https://doi.org/10.7910/DVN/9BCWKN

Creator	Kosuke Imai Dustin Tingley

Publisher	Harvard Dataverse

Description	Empirical testing of competing theories lies at the heart of social science research. We demonstrate that a well-known class of statistical models, called ﬁnite mixture models, provides an effective way of rival theory testing. In the proposed framework, each observation is assumed to be generated from a statistical model implied by one of the competing theories or more generally from a weighted combination of multiple statistical models under consideration. Researchers can then estimate the probability that a speciﬁc observation is consistent with either of the competing theories. By directly modeling this probability with covariates, one can also determine the conditions under which a particular theory applies. We discuss a principled way to identify a list of observations that are statistically signiﬁcantly consistent with each theory. Finally, we propose several measures of the overall performance of each rival theory. We illustrate the advantages of our method through empirical and simulation studies.

Subject	finite mixture models, theory testing, non-nested model selection

Date	2011

ICAR Research Data Repository for Knowledge Management