Forecasts in a Slightly Misspecified Finite Order VAR∗
نویسندگان
چکیده
We propose a Bayesian procedure for exploiting small, possibly long-lag linear predictability in the innovations of a finite order autoregression. We model the innovations as having a log-spectral density that is a continuous mean-zero Gaussian process of order 1/ √ T . This local embedding makes the problem asymptotically a normal-normal Bayes problem, resulting in closed-form solutions for the best forecast. When applied to data on 132 U.S. monthly macroeconomic time series, the method is found to improve upon autoregressive forecasts by an amount consistent with the theoretical and Monte Carlo calculations. JEL classification: C53; C32; C11
منابع مشابه
An Evaluation of Cbo Forecasts
We compare the performance of a subset of CBO’s economic forecasts against that of an unrestricted vector autoregression (VAR) model. We evaluate forecasts of real economic indicators as well as budget-related nominal statistics. We find that under most specifications, the VAR performs competitively with, if slightly worse than, the corresponding CBO forecasts at up to 20 quarters. Therefore, a...
متن کاملFour-dimensional variational data assimilation with a diabatic version of the NCEP global spectral model: System development and preliminary results
The formulation of the National Centers for Environmental Prediction four-dimensional variational dataassimilation (4D-Var) system is described. Results of applying 4D-Var over a one-week assimilation period, with a full set of physical parametrizations, are presented and compared with those of 3D-Var. The linearization has been performed without simplifications and, therefore, the tangent-line...
متن کاملAn Alternative Explanation of the Price Puzzle
This paper proposes a simple explanation for the frequent appearance of a price puzzle in VARs designed for monetary policy analysis. It suggests that the best method of solving the puzzle implies a close connection between theory and empirics rather than the introduction of a commodity price. It proves that the omission of a measure of output gap (or potential output) spuriously produces a pri...
متن کاملVAR Priors : Success or lack of a decent macroeconomic theory ?
The purpose of this paper is to demonstrate that the success of the Litterman prior in VAR forecasting is not due to the realism of the prior, but rather because the prior conveniently reduces forecast error variance in common cases of misspecification. Specifically, it is shown that the imposition of a random walk prior reduces forecast error variance in misspecifications involving (1) time-va...
متن کاملDSGE model restrictions for structural VAR identification
The identification of reduced-form VAR model had been the subject of numerous debates in the literature. Different sets of identifying assumptions can lead to very different conclusions in the policy debate. This paper proposes a theoretical consistent identification strategy using restrictions implied by a DSGE model. Monte Carlo simulations suggest the proposed identification strategy is succ...
متن کامل