Kernel-based nonlinear canonical analysis and time reversibility

Serge Darolles; Jean Pierre Florens; Christian Gouriéroux

doi:10.1016/S0304-4076(03)00199-4

Kernel-based nonlinear canonical analysis and time reversibility

Serge Darolles
, Jean Pierre Florens
, Christian Gouriéroux

Research output: Contribution to journal › Article › peer-review

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.

Original language	English
Pages (from-to)	323-353
Number of pages	31
Journal	Journal of Econometrics
Volume	119
Issue number	2
DOIs	https://doi.org/10.1016/S0304-4076(03)00199-4
Publication status	Published - 1 Apr 2004
Externally published	Yes

Keywords

Diffusion equations
High-frequency data
Kernel estimators
Nonlinear canonical analysis
Reversibility hypothesis

Access to Document

10.1016/S0304-4076(03)00199-4

Cite this

@article{f1e9f231e7c440fc92408e1bb62a2d2e,

title = "Kernel-based nonlinear canonical analysis and time reversibility",

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, High-frequency data, Kernel estimators, Nonlinear canonical analysis, Reversibility hypothesis",

author = "Serge Darolles and Florens, \{Jean Pierre\} and Christian Gouri{\'e}roux",

year = "2004",

month = apr,

day = "1",

doi = "10.1016/S0304-4076(03)00199-4",

language = "English",

volume = "119",

pages = "323--353",

journal = "Journal of Econometrics",

issn = "0304-4076",

number = "2",

}

TY - JOUR

T1 - Kernel-based nonlinear canonical analysis and time reversibility

AU - Darolles, Serge

AU - Florens, Jean Pierre

AU - Gouriéroux, Christian

PY - 2004/4/1

Y1 - 2004/4/1

N2 - 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.

AB - 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.

KW - Diffusion equations

KW - High-frequency data

KW - Kernel estimators

KW - Nonlinear canonical analysis

KW - Reversibility hypothesis

UR - https://www.scopus.com/pages/publications/1642309374

U2 - 10.1016/S0304-4076(03)00199-4

DO - 10.1016/S0304-4076(03)00199-4

M3 - Article

AN - SCOPUS:1642309374

SN - 0304-4076

VL - 119

SP - 323

EP - 353

JO - Journal of Econometrics

JF - Journal of Econometrics

IS - 2

ER -

Kernel-based nonlinear canonical analysis and time reversibility

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this