Kernel Based Nonlinear Canonical Analysis
Serge Darolles, Jean‐Pierre Florens, Christian Gouriéroux · RePEc: Research Papers in Economics · 1999
We consider a kernel based approach to nonlinear canonical correlation analysis and its implementation for time series. We deduce various diagnostics for reversible processes and gaussian processes. The method is first applied to a stimulated series satisfying a diffusion equation allowing us to estimate nonparametrically the drift and volatility functions. The second application involves high frequency data on stock returns.