Abstract
In this paper we taken a rather different approach to the concept of cointegration (comparated to existing literature) by focusing on the distance norm of an appropriately defined stochastic process (the first differences of one series) and a closed linear subspace defined from the first differences of the other series. The main result contained in Theorem 2 states that, within a VAR(l) framework, two series are cointegrated if and only if this distance is smaller than the standard deviation of the former process. It links cointegration to the evaluation of the distance between two information sets concerning the short-run dynamic paths of the variables. Hence cointegration can be detected by the differenced series. We, also propose a test for cointegration
Similar content being viewed by others
Referencs
Engle, R. F., Granger, C. W. J. (1987). Cointegration and Error Correction: Representation. Estimation and Testing, Econometrica55, 251–76.
Granger, C. W. J. (1981). Some Properties of Time Series Data and Their Use in Econometric Model Specification, Journal of Econometrics16, 121–130.
Favero, C. A. (1996) Econometria, La Nuova Italia Scientifica, Roma.
Favero, C. A., Giavazzi, F., Spaventa, L. (1996). High Yields: The Spread on German Interest Rates, NBER Working Paper 5408.
Triacca, U. (2000). Cointegration and Distance between Information Sets. Econometric Theory16, 102–111.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Triacca, U. Cointegration in VAR(1) process: Characterization and testing. Statistical Papers 43, 435–443 (2002). https://doi.org/10.1007/s00362-002-0114-y
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/s00362-002-0114-y