Abstract

ABSTRACTThis paper reports a substantive application of Engle and Yoo's three‐step estimator for cointegrated systems. Their estimator was proposed as a computationally convenient alternative to a number of FIML systems estimators. In part this estimator was developed to overcome some drawbacks of the OLS estimator of the first stage Engle‐Granger cointegrating regression which, despite its widespread use, is not asymptotically efficient and does not provide (even) asympototically correct ‘t’ statistics. Our application, which is of interest in its own right, is to an explanation of expenditure on nondurables and services in the UK. In formulating an empirical explanation for this variable we find it necessary to extend our framework to consider a system of dynamic error correction equations with feedbacks — or error correction mechanisms — from consumption and housing equity withdrawal.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call