Non-parametric direct multi-step estimation for forecasting economic processes

被引：67

作者：

Chevillon, G ^{[1
]}

Hendry, DF

机构：

[1] Univ Oxford, Dept Econ, Oxford, England

[2] Observ Francais Conjonctures Econ, F-75340 Paris, France

[3] Univ Oxford, Nuffield Coll, Oxford, England

来源：

INTERNATIONAL JOURNAL OF FORECASTING | 2005年 / 21卷 / 02期

基金：

英国经济与社会研究理事会;

关键词：

adaptive estimation; multi-step estimation; dynamic forecasts; model mis-specification;

D O I：

10.1016/j.ijforecast.2004.08.004

中图分类号：

F [经济];

学科分类号：

02 ;

摘要：

We evaluate the asymptotic and finite-sample properties of direct multi-step estimation (DMS) for forecasting at several horizons. For forecast accuracy gains from DMS in finite samples, mis-specification and non-stationarity of the DGP are necessary, but when a model is well-specified, iterating the one-step ahead forecasts may not be asymptotically preferable. If a model is mis-specified for a non-stationary DGP, in particular omitting either negative residual serial correlation or regime shifts, DMS can forecast more accurately. Monte Carlo simulations clarify the nonlinear dependence of the estimation and forecast biases on the parameters of the DGP, and explain existing results. (c) 2004 International Institute of Forecasters. Published by Elsevier B.V. All rights reserved.

引用

页码：201 / 218

页数：18

共 40 条

[1]

Allen PG, 2001, INT SER OPER RES MAN, V30, P303

[2]

Banerjee A, 1996, OXFORD B ECON STAT, V58, P573

[3]

Bhansali RJ, 1997, STAT SINICA, V7, P425

[4]

Bhansali RJ, 1999, STAT TEXTB MONOG, V158, P201

[5] Asymptotically efficient autoregressive model selection for multistep prediction [J].

Bhansali, RJ .

ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS, 1996, 48 (03) :577-602

[6]

Box G, 1976, TIME SERIES ANAL FOR

[7]

Clements MP, 1996, OXFORD B ECON STAT, V58, P657

[8]

Clements MP, 1998, INT J FORECASTING, V14, P111, DOI 10.1016/S0169-2070(97)00057-5

[9]

CLEMENTS MP, 1999, FORECASTING NONSTATI

[10]

COX DR, 1961, J ROY STAT SOC B, V23, P414

← 1 2 3 4 →