Abstract: We investigate the relationship between ownership structure of financial assets and nonfundamental risk. We define an asset to be fragile if it susceptible to non-fundamental trading shocks. An asset can be fragile because of concentrated ownership, or because its owners face correlated liquidity shocks, ie., they must buy or sell at the same time. Two assets are “cofragile” if their owners have correlated trading needs, even if the holdings of these owners do not directly overlap. We formalize this idea and apply it to the ownership of US stocks between 1990 and 2007. Consistent with our predictions, fragility strongly predicts future price volatility, and co-fragility predicts cross-stock return comovement.

Abstract: Bayesian inference from high-dimensional data involves the integration over a large number of model parameters. Accurate evaluation of such high-dimensional integrals raises a unique set of issues. These issues are illustrated using the exemplar of model selection for principal component analysis (PCA). A Bayesian model selection criterion, based on a Laplace approximation to the model evidence for determining the number of signal principal components present in a data set, has previously been show to perform well on various test data sets. Using simulated data we show that for d-dimensional data and small sample sizes, N, the accuracy of this model selection method is strongly affected by increasing values of d. By taking proper account of the contribution to the evidence from the large number of model parameters we show that model selection accuracy is substantially improved. The accuracy of the improved model evidence is studied in the asymptotic limit d ! ¥ at fixed ratio a = N=d, with a < 1. In this limit, model selection based upon the improved model evidence agrees with a frequentist hypothesis testing approach.

Abstract: Feature selection is the task of choosing a small subset of features that is sufficient to predict the target labels well. Here, instead of trying to directly determine which features are better, we attempt to learn the properties of good features. For this purpose we assume that each feature is represented by a set of properties, referred to as meta-features. This approach enables prediction of the quality of features without measuring their value on the training instances. We use this ability to devise new selection algorithms that can efficiently search for new good features in the presence of a huge number of features, and to dramatically reduce the number of feature measurements needed. We demonstrate our algorithms on a handwritten digit recognition problem and a visual object category recognition problem. In addition, we show how this novel viewpoint enables derivation of better generalization bounds for the joint learning problem of selection and classification, and how it contributes to a better understanding of the problem. Specifically, in the context of object recognition, previous works showed that it is possible to find one set of features which fits most object categories (aka a universal dictionary). Here we use our framework to analyze one such universal dictionary and find that the quality of features in this dictionary can be predicted accurately by its meta-features.