Title of article :
Kernel-based nonlinear canonical analysis and time reversibility
Author/Authors :
Darolles، نويسنده , , Serge and Florens، نويسنده , , Jean-Pierre and Gouriéroux، نويسنده , , Christian، نويسنده ,
Issue Information :
دوفصلنامه با شماره پیاپی سال 2004
Abstract :
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.
Keywords :
Diffusion equations , Nonlinear canonical analysis , High-frequency data , Kernel estimators , Reversibility hypothesis
Journal title :
Journal of Econometrics
Journal title :
Journal of Econometrics
