Functional cointegration: definition and nonparametric estimation

Banerjee, A.; Pitarakis, J.-V.

Abstract

We formally define a concept of functional cointegration linking the dynamics of two time series via a functional coefficient. This is achieved through the use of a concept of summability as an alternative to I(1)’ness which is no longer suitable under nonlinear dynamics. We subsequently introduce a nonparametric approach for estimating the unknown functional coefficients. Our method is based on a piecewise local least squares principle and is computationally simple to implement. We establish its consistency properties and evaluate its performance in finite samples. We subsequently illustrate its usefulness through an application that explores linkages between stock prices and dividends via a sentiment indicator.

Citation

Banerjee, A., & Pitarakis, J. (2014). Functional cointegration: definition and nonparametric estimation. Studies in Nonlinear Dynamics & Econometrics, 18(5), 507-520. https://doi.org/10.1515/snde-2013-0083

• Journal Article Type: Article
• Online Publication Date: Jan 4, 2014
• Publication Date: Dec 1, 2014
• Deposit Date: Jun 3, 2014
• Publicly Available Date: Aug 30, 2017
• Journal: Studies in Nonlinear Dynamics and Econometrics
• Print ISSN: 1081-1826
• Electronic ISSN: 1558-3708
• Publisher: De Gruyter
• Peer Reviewed: Peer Reviewed
• Volume: 18
• Issue: 5
• Pages: 507-520
• DOI: https://doi.org/10.1515/snde-2013-0083
• Public URL: https://durham-repository.worktribe.com/output/1460757