Bayesian VAR Forecasts Fail to Live Up to Their Promise

Bischoff, Charles W.; Belay, Halefom; Kang, In-Bong
July 2000
Business Economics;Jul2000, Vol. 35 Issue 3, p19
Academic Journal
Macroeconomic forecasts from vector autoregression (VAR) models, in which the data are combined with Bayesian prior distributions on the coefficients, gained great popularity in the 1980s. The prior distribution known as the "Minnesota Prior" was used to make forecasts starting in 1980. These forecasts seemed, for a time, to not only equal but even to surpass those of the consulting groups selling forecasts based on large, judgmentally adjusted econometric models. Using actual forecasts made by the group then called DRI between 1981 and mid- 1996, we find the forecasts based on the "Minnesota Prior" did not continue their early success, even when they are averaged with the DRI forecasts.


Related Articles

  • Vector Autoregressions: Forecasting and Reality. Robertson, John C.; Tallman, Ellis W. // Economic Review (07321813);1999 1st Quarter, Vol. 84 Issue 1, p4 

    Discusses the steps in producing real-time forecasts from a vector autoregression (VAR) model. Information on economic forecasting; Focus on model-based forecast; Discussion on the development of a VAR model for forecasting.

  • Comment and Discussion.  // Brookings Papers on Economic Activity;2001, Issue 1, p208 

    This article presents commentaries on a paper by Philip Howrey about the use of the Index of Consumer Sentiment (ICS) developed by George Katona in forecasting business cycles. Two things about the ICS deserve particular notice. First, Katona did not create the ICS for forecasting purposes, or...

  • Moment matrices in conditional heteroskedastic models under elliptical distributions with applications in AR-ARCH models. Liu, Shuangzhe; Heyde, Chris; Wong, Wing-Keung // Statistical Papers;Aug2011, Vol. 52 Issue 3, p621 

    It is well known that moment matrices play a very important rôle in econometrics and statistics. Liu and Heyde (Stat Pap 49:455-469, 2008) give exact expressions for two-moment matrices, including the Hessian for ARCH models under elliptical distributions. In this paper, we extend the theory...

  • Autoregression and Decision Making under Uncertainty. Shrivastava, Utkarsh; Prakash, Gyan; Dhar, Joydip; Omar, Arti // Modern Applied Science;Aug2010, Vol. 4 Issue 8, p144 

    There are innumerable social and economic situations in which we are influenced in our decision making by what others are doing. Under uncertainty it's general tendency of an individual to get inspired by decisions of others or mass opinion. However, such herd behavior many times leads to...

  • BMI Methodology.  // Turkey Defence & Security Report;Q2 2011, Issue 2, p93 

    The article offers information on how Business Monitor International Ltd. (BMI) generates industry forecasts for the defence and security industry of Turkey. It states that BMI uses autoregressive moving average (ARMA) model and vector autoregressions which allow them to forecast a variable...

  • Mexico's balance of payments and exchange rates: A cointegration analysis. De La Cruz Martinez, Justino // North American Journal of Economics & Finance;1999, Vol. 10 Issue 2, p401 

    Mexico's recurrent economic crises have cast serious doubts on the existence of a long-run relationship between the country's balance-of-payments and exchange rates. In this paper, cointegration and vector autoregression techniques are applied to Mexico's data covering the period 1971 through...

  • The Search Algorithm of Optimal Time Series Model for Forecasting Latvian GDP. Bezrucko, Aleksandrs // Computer Science (1407-7493);2010, Vol. 42, p83 

    No abstract available.

  • Prior Parameter Uncertainty: Some Implications for Forecasting and Policy Analysis with VAR Models. Robertson, John C.; Tallman, Ellis W. // Working Paper Series (Federal Reserve Bank of Atlanta);Oct99, Vol. 1999 Issue 13, p1 

    Examines how the treatment of prior uncertainty about parameter values can affect forecasting accuracy and the interpretation of identified vector autoregression (VAR) models. Improvement of VAR forecasting method after the use of informative stochastic prior restrictions; Impact of priors on...

  • On Forecasting With a First-Order Autoregression.  // Atlantic Economic Journal;Sep94, Vol. 22 Issue 3, p80 

    Previews the report `On Forecasting With a First-Order Autoregression,' by Jose Ramon Cancelo.


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics