Inflation, forecast intervals and long memory regression models

Charles S. Bos1, Philip Hans Franses1 and Marius Ooms2

4 April 2001

Abstract

We examine recursive out-of-sample forecasting of monthly postwar U.S. core inflation and log price levels. We use the autoregressive fractionally integrated moving average model with explanatory variables (ARFIMAX). Our analysis suggests a significant explanatory power of leading indicators associated with macroeconomic activity and monetary conditions for forecasting horizons up to two years. Correcting for the effect of explanatory variables, we still find fractional integration and structural breaks in the mean and variance of inflation in the 1970s and 1980s. We compare the forecasts of ARFIMAX models and ARIMAX models over the period 1984-1999. The ARIMAX(1,1,1) model provides the best forecasts, but its multi-step forecast intervals are too large. The multi-step forecast intervals of the ARFIMAX(0,d,0) model prove to be more realistic.

Keywords: Long Memory, inflation, time series, recursive estimation, multi-step forecasting.


Footnotes:

1Econometric Institute, Erasmus University, Rotterdam, and Tinbergen Institute, The Netherlands
2 Corresponding author, Department of Econometrics and Operations Research, vrije Universiteit amsterdam, De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands, email: mooms@econ.vu.nl


File translated from TEX by TTH, version 2.72.


[Main page] [Top] [Next: Marginal likelihood]
Last change: 7/12/2001