Professor of Economics - London Business School
On Persistence of Shocks to Economic Variables: A Common Misconception
Reichlin, L., Lippi, M.
Date Published: 01/02/1992
Empirical results on U.S. GNP have provided estimates of the Beveridge and Nelson's persistence measure (almost) always greater than unity when using ARIMA models and always smaller than unity when using UCARIMA models. This paper shows that a measure of persistence less than unity is a mathematical consequence of the definition of UCARIMA models and not an independent estimation result. Therefore, ARIMA and UCARIMA estimates of persistence cannot be compared on the same ground.
Reichlin, L., Lippi, M., "On persistence of shocks to economic variables: a common misconception", Journal of Monetary Economics, Volume 29, Issue 1, February 1992, pp. 87-93
Search Google Scholar for this entry