Macroeconomic forecasting for Australia using a large number of predictors

Research output: Contribution to journalArticleResearchpeer-review

Abstract

A popular approach to forecasting macroeconomic variables is to utilize a large number of predictors. Several regularization and shrinkage methods can be used to exploit such high-dimensional datasets, and have been shown to improve forecast accuracy for the US economy. To assess whether similar results hold for economies with different characteristics, an Australian dataset containing observations on 151 aggregate and disaggregate economic series as well as 185 international variables, is introduced. An extensive empirical study is carried out investigating forecasts at different horizons, using a variety of methods and with information sets containing an increasing number of predictors. In contrast to other countries the results show that it is difficult to forecast Australian key macroeconomic variables more accurately than some simple benchmarks. In line with other studies we also find that there is little to no improvement in forecast accuracy when the number of predictors is expanded beyond 20–40 variables and international factors do not seem to help.

Original languageEnglish
Pages (from-to)616-633
Number of pages18
JournalInternational Journal of Forecasting
Volume35
Issue number2
DOIs
Publication statusPublished - 1 Apr 2019

Keywords

  • Bagging
  • Bayesian VAR
  • Dynamic factor model
  • Least angular regression
  • Regularization
  • Ridge regression
  • Shrinkage

Cite this

@article{c9b5784dd0e84cc3bcbae43a7ac555ac,
title = "Macroeconomic forecasting for Australia using a large number of predictors",
abstract = "A popular approach to forecasting macroeconomic variables is to utilize a large number of predictors. Several regularization and shrinkage methods can be used to exploit such high-dimensional datasets, and have been shown to improve forecast accuracy for the US economy. To assess whether similar results hold for economies with different characteristics, an Australian dataset containing observations on 151 aggregate and disaggregate economic series as well as 185 international variables, is introduced. An extensive empirical study is carried out investigating forecasts at different horizons, using a variety of methods and with information sets containing an increasing number of predictors. In contrast to other countries the results show that it is difficult to forecast Australian key macroeconomic variables more accurately than some simple benchmarks. In line with other studies we also find that there is little to no improvement in forecast accuracy when the number of predictors is expanded beyond 20–40 variables and international factors do not seem to help.",
keywords = "Bagging, Bayesian VAR, Dynamic factor model, Least angular regression, Regularization, Ridge regression, Shrinkage",
author = "Anastasios Panagiotelis and George Athanasopoulos and Hyndman, {Rob J.} and Bin Jiang and Farshid Vahid",
year = "2019",
month = "4",
day = "1",
doi = "10.1016/j.ijforecast.2018.12.002",
language = "English",
volume = "35",
pages = "616--633",
journal = "International Journal of Forecasting",
issn = "0169-2070",
publisher = "Elsevier",
number = "2",

}

Macroeconomic forecasting for Australia using a large number of predictors. / Panagiotelis, Anastasios; Athanasopoulos, George; Hyndman, Rob J.; Jiang, Bin; Vahid, Farshid.

In: International Journal of Forecasting, Vol. 35, No. 2, 01.04.2019, p. 616-633.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Macroeconomic forecasting for Australia using a large number of predictors

AU - Panagiotelis, Anastasios

AU - Athanasopoulos, George

AU - Hyndman, Rob J.

AU - Jiang, Bin

AU - Vahid, Farshid

PY - 2019/4/1

Y1 - 2019/4/1

N2 - A popular approach to forecasting macroeconomic variables is to utilize a large number of predictors. Several regularization and shrinkage methods can be used to exploit such high-dimensional datasets, and have been shown to improve forecast accuracy for the US economy. To assess whether similar results hold for economies with different characteristics, an Australian dataset containing observations on 151 aggregate and disaggregate economic series as well as 185 international variables, is introduced. An extensive empirical study is carried out investigating forecasts at different horizons, using a variety of methods and with information sets containing an increasing number of predictors. In contrast to other countries the results show that it is difficult to forecast Australian key macroeconomic variables more accurately than some simple benchmarks. In line with other studies we also find that there is little to no improvement in forecast accuracy when the number of predictors is expanded beyond 20–40 variables and international factors do not seem to help.

AB - A popular approach to forecasting macroeconomic variables is to utilize a large number of predictors. Several regularization and shrinkage methods can be used to exploit such high-dimensional datasets, and have been shown to improve forecast accuracy for the US economy. To assess whether similar results hold for economies with different characteristics, an Australian dataset containing observations on 151 aggregate and disaggregate economic series as well as 185 international variables, is introduced. An extensive empirical study is carried out investigating forecasts at different horizons, using a variety of methods and with information sets containing an increasing number of predictors. In contrast to other countries the results show that it is difficult to forecast Australian key macroeconomic variables more accurately than some simple benchmarks. In line with other studies we also find that there is little to no improvement in forecast accuracy when the number of predictors is expanded beyond 20–40 variables and international factors do not seem to help.

KW - Bagging

KW - Bayesian VAR

KW - Dynamic factor model

KW - Least angular regression

KW - Regularization

KW - Ridge regression

KW - Shrinkage

UR - http://www.scopus.com/inward/record.url?scp=85061617145&partnerID=8YFLogxK

U2 - 10.1016/j.ijforecast.2018.12.002

DO - 10.1016/j.ijforecast.2018.12.002

M3 - Article

VL - 35

SP - 616

EP - 633

JO - International Journal of Forecasting

JF - International Journal of Forecasting

SN - 0169-2070

IS - 2

ER -