Long Memory in Inflation Rates: International Evidence
- 1 January 1995
- journal article
- research article
- Published by Taylor & Francis in Journal of Business & Economic Statistics
- Vol. 13 (1) , 37-45
- https://doi.org/10.1080/07350015.1995.10524577
Abstract
We examine monthly inflation rates of five industrial countries. The application of tests against stationarity as well as tests against a unit root yield contradictory results. Thus fractional integration allowing for long memory is a plausible model. We discuss and apply the periodogram regression to estimate the difference parameters. For all countries we find estimates significantly different from 1 as well as from 0. This is evidence in favor of long memory. Specification tests and maximum likelihood estimates support the fitted models. Finally, we relate our empirical results to the construction of the data.Keywords
This publication has 32 references indexed in Scilit:
- On maximum likelihood estimation of the differencing parameter of fractionally-integrated noise with unknown meanJournal of Econometrics, 1994
- BIAS IN AN ESTIMATOR OF THE FRACTIONAL DIFFERENCE PARAMETERJournal of Time Series Analysis, 1993
- On the power of Dickey-Fuller tests against fractional alternativesEconomics Letters, 1991
- The Fisher hypothesis and the forecastability and persistence of inflationJournal of Monetary Economics, 1987
- Large-Sample Properties of Parameter Estimates for Strongly Dependent Stationary Gaussian Time SeriesThe Annals of Statistics, 1986
- THE ESTIMATION AND APPLICATION OF LONG MEMORY TIME SERIES MODELSJournal of Time Series Analysis, 1983
- Likelihood Ratio Statistics for Autoregressive Time Series with a Unit RootEconometrica, 1981
- Long memory relationships and the aggregation of dynamic modelsJournal of Econometrics, 1980
- AN INTRODUCTION TO LONG‐MEMORY TIME SERIES MODELS AND FRACTIONAL DIFFERENCINGJournal of Time Series Analysis, 1980
- The Typical Spectral Shape of an Economic VariableEconometrica, 1966