Improving Economic Forecasting With Bayesian Vector Autoregression

Todd, Richard M.

doi:10.21034/qr.843

Cited by 101 publications

(56 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Two versions of the BVAR prior restrictions were employed, one reflecting the Minnesota prior, and the other employing a block recursive form of prior weighting restrictions. The Minnesota prior uses a prior overall tightness of 0.1, a harmonic lag decay of 0.1, weights of unity for the lags of the own variable in each equation, and symmetric weights of 0.5 for the lagged variables from other equations (Litterman, 1984;Todd, 1984). The recursive prior used a prior overall tightness of 0.2, harmonic lag decay of 0.1, weights of unity for the lags of the own variable in each equation, weights of 0.1 for the lower triangular part of the weighting matrix representing the influence of export employment on local employment, and weights of unity for the upper triangular part reflecting causality flowing from export employment to local employment.…”

Section: Forecasting Experiments With the Modelmentioning

confidence: 99%

Forecasting Metropolitan Employment Using an Export‐base Error‐corbection Model

LeSage

1990

Journal of Regional Science

View full text Add to dashboard Cite

ABSTRGCT. This paper shows that a time-series version of export-based location quotients, which decompose area employment into export and local components are co-integrated. Given the statistically established co-integrating relation between the time series for export and local employment, the Granger Representation Theorem tells us that an error-correction mechanism model is the appropriate time-series model specification. A set of forecasting experiments is carried out, in order to test whether the error-correction model produces better long-run forecasts that hold together in economically meaningful ways. Given the computer hardware and software that currently exist, the model developed here represents a simple, low-cost method for accurately forecasting local-area employment in the tradition of past work on export-base models.

show abstract

Section: Forecasting Experiments With the Modelmentioning

confidence: 99%

Forecasting Metropolitan Employment Using an Export‐base Error‐corbection Model

LeSage

1990

Journal of Regional Science

View full text Add to dashboard Cite

show abstract

“…The Bayesian VAR, as described in Litterman (1981), Doan et al (1984), Todd (1984), Litterman (1986) and Spencer (1993), has become a widely popular approach to dealing with overparametrization. One of main problems in using VAR models is that many parameters need to be estimated, although some of them may be insignificant.…”

Section: Bayesian Varmentioning

confidence: 99%

Policy‐Oriented Macroeconomic Forecasting with Hybrid DGSE and Time‐Varying Parameter VAR Models

Bekiros

Paccagnini

2016

Journal of Forecasting

View full text Add to dashboard Cite

Micro-founded dynamic stochastic general equilibrium (DSGE) models appear to be particularly suited for evaluating the consequences of alternative macroeconomic policies. Recently, increasing e¤orts have been undertaken by policymakers to use these models for forecasting, although this proved to be problematic due to estimation and identi…cation issues. Hybrid DSGE models have become popular for dealing with some of model misspeci…cations and the trade-o¤ between theoretical coherence and empirical …t, thus allowing them to compete in terms of predictability with VAR models. However, DSGE and VAR models are still linear and they do not consider time-variation in parameters that could account for inherent nonlinearities and capture the adaptive underlying structure of the economy in a robust manner. This study conducts a comparative evaluation of the out-of-sample predictive performance of many di¤erent speci…cations of DSGE models and various classes of VAR models, using datasets for the real GDP, the harmonized CPI and the nominal short-term interest rate series in the Euro area. Simple and hybrid DSGE models were implemented including DSGE-VAR and Factor Augmented DGSE, and tested against standard, Bayesian and Factor Augmented VARs. Moreover, a new state-space time-varying VAR model is presented. The total period spanned from 1970:1 to 2010:4 with an out-of-sample testing period of 2006:1-2010:4, which covers the global …nancial crisis and the EU debt crisis. The results of this study can be useful in conducting monetary policy analysis and macro-forecasting in the Euro area.JEL Classi…cation: C11, C15, C32

show abstract

“…Given that VAR models contain no exogenous variables, they are typically used to generate 'unconditional forecasts', which do not depend on explicit assumptions about the future path of exogenous variables. Moreover, due to the complete generality with which they specify the intercorrelations between the variables, VARs have been found particularly successful for this purpose, especially with a Bayesian treatment of parameter estimation (see Litterman, 1984;Todd, 1984). Since VAR models rely mainly on regularities in the historical data, the implicit assumption on which the forecasts are based is that the future behaviour of all variables will be the same as in the past.…”

Section: Unconditional Forecastsmentioning

confidence: 99%

“…This approach combines data evidence with prior knowledge about the likely value of the coefficients, supplied in the form of Bayesian priors. A number of studies suggest that the Bayesian vector autoregression (BVAR) technique improves the forecasting performance of large unrestricted VAR (UVAR) models and compares favorably with other forecasting procedures (see Litterman, 1984;Todd, 1984;Doan et al, 1984). The BVAR specification I Model-building activity in Italy has not been as widespread as in other countries.…”

Section: Description Of the Modelsmentioning

confidence: 99%

Comparing ex‐ante forecasts from a sem and var model: An application to the italian economy

Boero

1990

Journal of Forecasting

View full text Add to dashboard Cite

The predictive performance of a large-scale structural econometric model (SEM) of the Italian economy-the Prometeia model-is compared in this paper with a vector autoregressive (VAR) model estimated for a selection of six main variables of interest. The paper concentrates on the quarterly ex-ante forecasts of GDP growth rate and the annual forecasts of GDP growth and inflation rate, over the period 1980-85. It concludes that no forecaster is systematically better than the other. In particular, the VAR model outperforms the SEM in short-run forecasts, suggesting that, for the latter, more careful attention should be addressed to questions of dynamic specification. On the other hand, for longer intervals, the SEM forecasts are more accurate than the VAR forecasts, in that they can benefit from the judgemental interventions of the model users and the model can pick up the non-linearities of the economy which cannot be captured by the VAR. Given the different kinds of information that can be extracted from the two approaches, it seems more reasonable to consider them as complementary rather than alternative tools for modelling and forecasting. Therefore, rather than attempting to establish the superiority of one type of model over the other, this kind of comparisons should be seen as a useful diagnostic tool for detecting types of model misspecification. KEY WORDS Structural models Vector autoregressive models Ex-ante forecasts Comparison of forecasts RMSEsSubstantial work has been done in comparing ex-ante forecasts of large-scale structural models with time series forecasts. Most of the analysis has been addressed to UK and US models, using as a common benchmark of comparison univariate ARIMA models of the Box-Jenkins type (Holden et al., 1982; Longbottom and Holly, 1985; McNees, 1979 McNees, , 1982. A common finding from these studies is that time series forecasts can be good competitors for SEM forecasts in the short run but not over long horizons.

show abstract

Improving Economic Forecasting With Bayesian Vector Autoregression

Cited by 101 publications

References 5 publications

Forecasting Metropolitan Employment Using an Export‐base Error‐corbection Model

Forecasting Metropolitan Employment Using an Export‐base Error‐corbection Model

Policy‐Oriented Macroeconomic Forecasting with Hybrid DGSE and Time‐Varying Parameter VAR Models

Comparing ex‐ante forecasts from a sem and var model: An application to the italian economy

Contact Info

Product

Resources

About