No Detail Found!

Bootstrap forecasts versus conventional forecasts: a comparison for a second order autoregressive model.

Author(s)
Mun S. Son Mathematics & Statistics Department, University Vermont, Burlington, USA
Hosny I. Hamdy Mathematics & Statistics Department, University Vermont, Burlington, USA
Donald Holbert School of Allied Health, East Carolina University, Greenville, USA
Abstract
Multi-period forecasts and their associated standard errors are generated for a particular AR(2) model using two fundamentally different approaches: (1) The widely accepted methods outlined in Box and Jenkins (1976). (2) The relatively more recent method of bootstrapping, introduced by Efron (1979). For the discrete error distribution used in this study, the conventional Box-Jenkins methods appear to be conservative.
Publication Details
Page(s) 33-44
DOI DOI not available
Published Journal: Pakistan Journal of Statistics, Volume: 8, Issue: 1, Year: 1992

Request the full text from PASTIC

More from Pakistan Journal of Statistics
0

Citations

0

Downloads

45

Views

Copyright ©  PASTIC National Center, Islamabad www.pastic.gov.pk