Statistical Modelling 10 (2010), 379–390

Bayesian outlier detection in Capital Asset Pricing Model

Maria Elena De Giuli
Department of Economics and Quantitative Methods,
University of Pavia
Italy

Mario Alessandro Maggi
Department of Economics and Quantitative Methods,
University of Pavia
Italy

Claudia Tarantola
Department of Economics and Quantitative Methods,
Via S. Felice 7,
I–27100 Pavia
Italy
eMail: claudia.tarantola@unipv.it

Abstract:

We propose a novel Bayesian optimization procedure for outlier detection in the Capital Asset Pricing Model. We use a parametric product partition model to robustly estimate the systematic risk of an asset. We assume that the returns follow independent normal distributions and we impose a partition structure on the parameters of interest. The partition structure imposed on the parameters induces a corresponding clustering of the returns. We identify via an optimization procedure the partition that best separates standard observations from the atypical ones. The methodology is illustrated with reference to a real dataset, for which we also provide a microeconomic interpretation of the detected outliers.

Keywords:

Capital Asset Pricing Model; constrained optimization algorithm; Markov chain Monte Carlo; outlier identification; parametric product partition models; score function
back