Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Hurvich, Clifford M. | - |
dc.date.accessioned | 2006-06-22T17:53:19Z | - |
dc.date.available | 2006-06-22T17:53:19Z | - |
dc.date.issued | 2000 | - |
dc.identifier.uri | http://hdl.handle.net/2451/14794 | - |
dc.description.abstract | We develop forecasting methodology for the fractional exponential (FEXP) model. First, we devise algorithms for fast exact computation of the coefficients in the infinite order autoregressive and moving average representations of a FEXP process. We also describe an algorithm to accurately approximate the autocovariances and to simulate realizations of the process. Next, we present a fast frequency-domain cross validation method for selecting the order of the model. This model selection method is designed to yield the model which provides the best multistep forecast for the given lead time, without assuming that the process actually obeys a FEXP model. Finally, we use the infinite order autoregressive coefficients of a fitted FEXP model to construct multistep forecasts of inflation in the United Kingdom. These forecasts are substantially different than those from a fitted ARFIMA model. | en |
dc.format.extent | 191796 bytes | - |
dc.format.mimetype | application/pdf | - |
dc.language | English | EN |
dc.language.iso | en | |
dc.publisher | Stern School of Business, New York University | en |
dc.relation.ispartofseries | SOR-2000-10 | en |
dc.subject | Fractional integration | en |
dc.subject | Long-range dependence | en |
dc.subject | Spectral factorization | en |
dc.title | Multistep forecasting of long memory series using fractional exponential models | en |
dc.type | Working Paper | en |
dc.description.series | Statistics Working Papers Series | EN |
Appears in Collections: | IOMS: Statistics Working Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
SOR-2000-10.pdf | 187.3 kB | Adobe PDF | View/Open |
Items in FDA are protected by copyright, with all rights reserved, unless otherwise indicated.