Thumbnail

Kernel-based nonlinear canonical analysis and time reversibility

Florens, Jean-Pierre; Gouriéroux, Christian; Darolles, Serge (2004), Kernel-based nonlinear canonical analysis and time reversibility, Journal of Econometrics, 119, 2, p. 323–353. http://dx.doi.org/10.1016/S0304-4076(03)00199-4

View/Open
Type
Article accepté pour publication ou publié
Date
2004-04
Journal name
Journal of Econometrics
Volume
119
Number
2
Publisher
Elsevier
Pages
323–353
Publication identifier
Metadata
Show full item record
Author(s)
Florens, Jean-Pierre
Gouriéroux, Christian
Darolles, Serge
Abstract (EN)
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.
Subjects / Keywords
Nonlinear canonical analysis; Kernel estimators; Reversibility hypothesis; Diffusion equations; High-frequency data
JEL
C14 - Semiparametric and Nonparametric Methods: General
C22 - Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes
C40 - General