Leaders and followers in mutual funds: A dynamic Bayesian approach

This article proposes a dynamic Bayesian framework to analyze the leadership relationships between mutual funds. To this end, a two‐step procedure is proposed. First, a Bayesian rolling window based on the Capital Asset Pricing Model is used to estimate the evolution of mutual funds' market exposure over time. Then, a vector autoregressive (VAR) model is used to analyze the leader‐follower relationship between pair of mutual funds. Several leadership measures are studied. An application to Spanish mutual funds is carried out. In addition, the study examines the determining factors of mutual fund leadership.     

Dynamic style analysis of Spanish balanced pension plans: A Bayesian approach

This paper is focused on the dynamic allocations of Spanish balanced pension plans that invest predominantly in Euro‐zone equities. Applying a Bayesian method to a return‐based style analysis that includes the constraints of the strong version and time‐varying exposures, we provide evidence for no statistically significant changes over time in the main strategic asset allocations, namely, equity assets, long‐term debt and cash allocations. However, we find time‐varying selection abilities, indicating that the value added by managers is not the same over time. Although the investment style tends to be constant in each pension plan, these allocations are variable across plans which allow us to find different subsets of portfolios that present different mean returns and volatilities. Some pension plan features, such as size and type of financial institution that manages the portfolio, have been considered in trying to find concurrent characteristics in each subset. Copyright © 2010 John Wiley & Sons, Ltd.     

Bayesian analysis of mixture of autoregressive components with an application to financial market volatility

In this paper, we present a fully Bayesian analysis of a finite mixture of autoregressive components. Neither the number of mixture components nor the autoregressive order of each component have to be fixed, since we treat them as stochastic variables. Parameter estimation and model selection are performed using Markov chain Monte Carlo methods. This analysis allows us to take into account the stationarity conditions on the model parameters, which are often ignored by Bayesian approaches. Finally, the application to return volatility of financial markets will be illustrated. Our model seems to be consistent with some empirical facts concerning volatility such as persistence, clustering effects, nonsymmetrical dependencies. Copyright © 2006 John Wiley & Sons, Ltd.