Estimation of a Noninvertible Moving Average Process

The Case of Overdifferencing


Charles I. Plosser

President, Federal Reserve Bank of Philadelphia

G. William Schwert

University of Rochester, Rochester, NY 14627
and National Bureau of Economic Research


Journal of Econometrics, 6 (September 1977) 199-224


The effect of differencing all of the variables in a properly specified regression equation is examined. Excessive use of the difference transformation induces a non-invertible moving average (MA) process in the disturbances of the transformed regression. Monte Carlo technniques are used to examine the effects of overdifferencing on the efficiency of regression parameter estimates, inferences based on those estimates, and tests for overdifferencing based on the estimator of the MA parameter for the disturbances of the differences regression. Overall, the problem of overdifferencing is not serious if careful attention is paid to the properties of the disturbances of the regression equations.

Key words: Overdifferencing, ARIMA, Unit root

JEL Classifications: C22


Cited 75 times in the SSCI and SCOPUS through 2016
© Copyright 1977, Elsevier
The following file contains the reprint of this paper in Acrobat's portable data format (.pdf). The file is about 1,205 KB and can only be viewed (and printed) using a copy of Acrobat Reader or Acrobat Exchange.

If you want the current version of the Adobe Acrobat Reader for other platforms, visit Adobe's web page by clicking the image below.

Click here to download this paper in PDF format.


Return to Publications Page

© Copyright 1998-2017, G. William Schwert

Last Updated on 8/8/2017