Modeling Market Reactions to Auditor Changes Using Variable Selection Algorithms: A Meta-Analysis

Market reactions to auditor change filings have been studied over a long period in the literature. We provide a review of the literature on market response to auditor changes and identify a superset of variables used in published research. Applying methods from machine learning to optimize variable selection, we build models that explain market reaction to auditor changes. We compare the performance of our models with the performance of the models that use subsets of variables examined in a select list of studies in the literature. Our meta-analysis results in an improvement in model fit compared to the analysis used in prior studies.