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).

In this note we propose a general testing procedure for parametric models based on Bartlett Identities. A well-known example is the Information Matrix test, which is based on the Bartlett Identity of order 1. The Identities are shown to induce a sequenceof testable restrictions on the data...

In the case of panel data, the introduced explanatory variables are often not sufficient for describing all the individual caracteristics and it may be useful to consider relations with individual varying coefficients. When this heterogeneity is even partially omitted, the estimated parameters...

In this paper we are interested in inference problems on the matrix of coefficients in a multivariate linear model; in particular we consider tests on the kernel, the range and the rank of this matrix. Various test procedures are explicited and compared: (pseudo) likelihood ratio, Wald (or...