April 2008
This paper describes a framework for Bayesian multiple-hypothesis testing of autoregressive time series. Nonparametric characterizations of both the null and alternative hypotheses will be shown to be the key robustification step necessary to ensure reasonable Type-I error performance. Issues of prior specification are considered in detail. The methodology is applied to part of a large database containing up to 50 years of corporate performance statistics on 24,157 publicly traded American companies; the primary goal of the analysis is to flag companies whose historical performance is significantly different from that expected due to chance.
Keywords: multiple comparisons; Bayesian model selection; nonparametric Bayes; corporate performance.
The manuscript is in available in PDF format.
Cite as:
@Article{Scott:2009,
Author = "Scott, James G.",
Title = "Nonparametric Bayesian multiple testing for longitudinal performance stratification",
Year = 2009,
journal = {The Annals of Applied Statistics},
note = {to appear}
}