Interpreting cointegrating vectors and common stochastic trends

M R Wickens

Interpreting cointegrating vectors and common stochastic trends

Economics

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

Abstract

It often proves difficult to give a satisfactory economic interpretation to estimated cointegrating vectors derived from the maximum likelihood estimation of unrestricted vector correction models. This paper shows that this is because, without introducing a priori information, they are not identified. This is even true where there is only a single cointegrating vector. It is also shown that the common stochastic trends derived using VAR analysis in the presence of cointegration are not identified, nor can they be obtained uniquely from the estimated cointegrating vectors. The implication is that cointegration analysis needs to take account of structural restrictions after all. Consequently it is likely to be of less use in econometric model building than was first thought.

Original language	English
Pages (from-to)	255-271
Number of pages	17
Journal	Journal of Econometrics
Volume	74
Issue number	2
Publication status	Published - Oct 1996

Keywords

cointegration
common stochastic trends
econometrics
identification
nonstationarity
DEMAND
EXOGENEITY
REGRESSION
INFERENCE
SYSTEMS
MONEY
UK

Cite this

@article{41ab87716e3c489bb09363fd826a0068,

title = "Interpreting cointegrating vectors and common stochastic trends",

abstract = "It often proves difficult to give a satisfactory economic interpretation to estimated cointegrating vectors derived from the maximum likelihood estimation of unrestricted vector correction models. This paper shows that this is because, without introducing a priori information, they are not identified. This is even true where there is only a single cointegrating vector. It is also shown that the common stochastic trends derived using VAR analysis in the presence of cointegration are not identified, nor can they be obtained uniquely from the estimated cointegrating vectors. The implication is that cointegration analysis needs to take account of structural restrictions after all. Consequently it is likely to be of less use in econometric model building than was first thought.",

keywords = "cointegration, common stochastic trends, econometrics, identification, nonstationarity, DEMAND, EXOGENEITY, REGRESSION, INFERENCE, SYSTEMS, MONEY, UK",

author = "Wickens, {M R}",

year = "1996",

month = oct,

language = "English",

volume = "74",

pages = "255--271",

journal = "Journal of Econometrics",

issn = "0304-4076",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - Interpreting cointegrating vectors and common stochastic trends

AU - Wickens, M R

PY - 1996/10

Y1 - 1996/10

N2 - It often proves difficult to give a satisfactory economic interpretation to estimated cointegrating vectors derived from the maximum likelihood estimation of unrestricted vector correction models. This paper shows that this is because, without introducing a priori information, they are not identified. This is even true where there is only a single cointegrating vector. It is also shown that the common stochastic trends derived using VAR analysis in the presence of cointegration are not identified, nor can they be obtained uniquely from the estimated cointegrating vectors. The implication is that cointegration analysis needs to take account of structural restrictions after all. Consequently it is likely to be of less use in econometric model building than was first thought.

AB - It often proves difficult to give a satisfactory economic interpretation to estimated cointegrating vectors derived from the maximum likelihood estimation of unrestricted vector correction models. This paper shows that this is because, without introducing a priori information, they are not identified. This is even true where there is only a single cointegrating vector. It is also shown that the common stochastic trends derived using VAR analysis in the presence of cointegration are not identified, nor can they be obtained uniquely from the estimated cointegrating vectors. The implication is that cointegration analysis needs to take account of structural restrictions after all. Consequently it is likely to be of less use in econometric model building than was first thought.

KW - cointegration

KW - common stochastic trends

KW - econometrics

KW - identification

KW - nonstationarity

KW - DEMAND

KW - EXOGENEITY

KW - REGRESSION

KW - INFERENCE

KW - SYSTEMS

KW - MONEY

KW - UK

M3 - Article

SN - 0304-4076

VL - 74

SP - 255

EP - 271

JO - Journal of Econometrics

JF - Journal of Econometrics

IS - 2

ER -