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.
Highlights
Forecasts of macroeconomic variables, including key indicators such as Gross Domestic Product (GDP) and inflation, are necessary inputs for government budget planning, central bank policy making and business decisions
The first major contribution of this paper is to introduce an extensive Australian macroeconomic data set comparable in size to that of the US, comprising 151 quarterly Australian macroeconomic variables which naturally divide into 12 categories of macroeconomic activity
Our primary interest here was to first and foremost assess the value added to forecasting key Australian macroeconomic variables by increasing the size of the information set
Summary
Forecasts of macroeconomic variables, including key indicators such as GDP and inflation, are necessary inputs for government budget planning, central bank policy making and business decisions. The use of time series approaches for macroeconomic forecasting gained impetus in the 1970s and 1980s as forecasts from univariate ARIMA models (Box & Jenkins, 1970) and vector autoregressions (VARs) (Sims, 1980) were shown to outperform structural macroeconomic models (for a discussion of this history see Diebold, 1997, and references therein). During this era, the information sets used to form forecasts typically contained only a small number of variables. Two of the earliest and most widely used examples are the US dataset containing 149 variables measured at a monthly frequency featured in Stock and Watson (2002)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.