A periodic long memory model for quarterly UK inflation
View PDF Version
Redirect to publisher's version
(publisher's version.url.txt, 50 bytes)
We consider an extension of the fractionally integrated ARIMA(0, d, 0) model for quarterly UK inflation, where we allow the fractional integration parameter d to vary with the season s. This periodic ARFIMA(0, d, 0) model does not only provide an informative in-sample description, it may also be useful for out-of-sample forecasting. The main result is that the integration parameter in the first two quarters is significantly larger than that in the last two quarters.
- time series analysis
- ooms /
- estimation results