We consider a kernel-based approach to nonlinear canonical correlation analysis and its implementation for time series. We deduce a test procedure of the reversibility hypothesis. The method is applied to the analysis of stochastic differential equation from high-frequency data on stock returns.

We decompose a stationary Markov process (Xt) as a linear combination of ARMA. These decompositions are deduced from a nonlinear canonical decomposition of the joint distribution of (Xt, Xt−1).

A characteristic of hedge funds is not only an active portfolio management, but also the allocation of portfolio performance between different accounts, which are the accounts for the external investors, an account for the management firm and a provision account. Despite a lack of transparency...

High frequency transaction prices exhibit two major cha racteristics: they are discrete in level and only exist at random transaction dates. In this paper, we seek to model transaction price dynamics, taking into account these two features. We specify the transaction price process as a Markov...

This paper reviews the fund ratings based on Sharpe performance measures. We define a battery of performance measures in a mean-variance framework. They differ by the information taken into account in their computation, but also by the potential use of the fund by the investor. We apply these...