Can forecasting performance be improved by considering the steady state? An application to Swedish inflation and interest rate

Abstract

This paper investigates whether the forecasting performance of Bayesian autoregressive and vector autoregressive models can be improved by incorporating prior beliefs on the steady state of the time series in the system. Traditional methodology is compared to the new framework-in which a mean-adjusted form of the models is employed-by estimating the models on Swedish inflation and interest rate data from 1980 to 2004. Results show that the out-of-sample forecasting ability of the models is practically unchanged for inflation but significantly improved for the interest rate when informative prior distributions on the steady state are provided. The findings in this paper imply that this new methodology could be useful since it allows us to sharpen our forecasts in the presence of potential pitfalls such as near unit root processes and structural breaks, in particular when relying on small samples. Copyright © 2008 John Wiley & Sons, Ltd.

Real-Time Density Forecasts from VARs with Stochastic Volatility

“…The BVARs with most variables in gap form and steady state priors (for simplicity, much of the discussion below simply refers to these models as BVARs with steady state priors) generally yield lower RMSEs than conventional BVARs. This finding is in line with evidence in Clark andMcCracken (2008, 2010) on the advantage of detrending and evidence in Adolfson, et al (2007), Beechey and Osterholm (2008), Osterholm (2008), andWright (2010) on the advantage of steady state priors. The advantage is most striking for 2-year ahead forecasts of inflation.…”

“…Villani (2009) develops a Bayesian estimator of a (constant variance) VAR with an informative prior on the steady state. Applications of the estimator in studies such as Adolfson, et al (2007), Beechey and Osterholm (2008), Osterholm (2008), and Wright (2010) have shown that the use of a prior on the steady state often improves the accuracy of point forecasts. In a methodological sense, this paper extends the estimator of Villani (2009) to include stochastic volatility.…”

Does Money Growth Granger Cause Inflation in the Euro Area? Evidence from Out-of-Sample Forecasts Using Bayesian VARs*

“…This methodology has been shown to improve forecasts relative to traditional BVARs – see, for example, Adolfson et al. (2007) and Österholm (2008)– and will be used as a forecasting tool in this paper.…”

Bayesian VARs: Specification Choices and Forecast Accuracy

“… Studies such as Clark (), Osterholm () and Wright () have shown that the performance of models with variables transformed for stationarity can be improved significantly with the use of the steady‐state prior developed in Villani (2009). As Villani's estimation approach involves Gibbs sampling, it is significantly more computationally demanding than a model specified to permit a Normal‐inverted Wishart prior and posterior.…”