THE BEHAVIOR OF FORECAST ERRORS FROM A NEARLY INTEGRATED AR(1) MODEL AS BOTH SAMPLE SIZE AND FORECAST HORIZON BECOME LARGE

DOI10.1017/S026646669915206XMaRDI QIDQ4512734zbMATH OpenOpenAlexFDO

Authors Gordon C. R. Kemp

Publication date 5 November 2000

Published in Econometric Theory (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1017/s026646669915206x

zbMATH Keywords

Monte Carlo studyAR(1) modelforecast errors

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Inference from stochastic processes and prediction (62M20)

Recommendations

• Finite-sample properties of forecasts from the stationary first-order autoregressive model under a general error distribution
• The adequacy of asymptotic approximations in the near-integrated autoregressive model with dependent errors
• Approximations to some exact distributions in the rrasr orderautoregressive model with dependenterrors
• Forecast Error Symmetry in ARIMA Models
• Asymptotic mean‐squared forecast error when an autoregression with linear trend is fitted to data generated by an I(0) or I(1) process

Cited in

(5)

• Median unbiased forecasts for highly persistent autoregressive processes
• Finite-sample properties of forecasts from the stationary first-order autoregressive model under a general error distribution
• Asymptotic mean‐squared forecast error when an autoregression with linear trend is fitted to data generated by an I(0) or I(1) process
• Nearly weighted risk minimal unbiased estimation
• Averaging estimators for autoregressions with a near unit root

This page was built for publication: THE BEHAVIOR OF FORECAST ERRORS FROM A NEARLY INTEGRATED AR(1) MODEL AS BOTH SAMPLE SIZE AND FORECAST HORIZON BECOME LARGE

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4512734)