In the existing literature, conditional forecasts in the vector autoregressive (VAR) framework have not been commonly presented with probability distributions. This paper develops Bayesian methods for computing the exact finite-sample distribution of conditional forecasts. It broadens the class of conditional forecasts to which the methods can be applied. The methods work for both structural and reduced-form VAR models and, in contrast to common practices, account for parameter uncertainty in finite samples. Empirical examples under both a flat prior and a reference prior are provided to show the use of these methods.

There has been a call for caution regarding the standard procedure for Bayesian inference in set‐identified structural vector autoregressions on the grounds that the common practice of using a uniform prior over the set of orthogonal matrices induces a non‐uniform prior for individual impulse responses or other quantities of interest. This paper challenges this call by formally showing that when the focus is on joint inference, the uniform prior over the set of orthogonal matrices is not only sufficient but also necessary for inference based on a uniform joint prior distribution over the identified set for the vector of impulse responses. In addition, we show how to conduct inference based on a uniform joint prior distribution for the vector of impulse responses.

In this paper, we develop algorithms to independently draw from a family of conjugate posterior distributions over the structural parameterization when sign and zero restrictions are used to identify structural vector autoregressions (SVARs). We call this family of conjugate posteriors normal‐generalized‐normal. Our algorithms draw from a conjugate uniform‐normal‐inverse‐Wishart posterior over the orthogonal reduced‐form parameterization and transform the draws into the structural parameterization; this transformation induces a normal‐generalized‐normal posterior over the structural parameterization. The uniform‐normal‐inverse‐Wishart posterior over the orthogonal reduced‐form parameterization has been prominent after the work of Uhlig (2005). We use Beaudry, Nam, and Wang's (2011) work on the relevance of optimism shocks to show the dangers of using alternative approaches to implementing sign and zero restrictions to identify SVARs like the penalty function approach. In particular, we analytically show that the penalty function approach adds restrictions to the ones described in the identification scheme.

We propose an approach for Bayesian inference in time-varying structural vector autoregressions (SVARs) identified with sign restrictions. The linchpin of our approach is a class of rotation-invariant time-varying SVARs in which the prior and posterior densities of any sequence of structural parameters belonging to the class are invariant to orthogonal transformations of the sequence. Our methodology is new to the literature. In contrast to existing algorithms for inference based on sign restrictions, our algorithm is the first to draw from a uniform distribution over the sequences of orthogonal matrices given the reduced-form parameters. We illustrate our procedure for inference by analyzing the role played by monetary policy during the latest inflation surge.

Structural vector autoregressions (SVARs) are widely used for policy analysis and to provide stylized facts for dynamic stochastic general equilibrium (DSGE) models; yet no workable rank conditions to ascertain whether an SVAR is globally identified have been established. Moreover, when nonlinear identifying restrictions are used, no efficient algorithms exist for small-sample estimation and inference. This paper makes four contributions towards filling these important gaps in the literature. First, we establish general rank conditions for global identification of both identified and exactly identified models. These rank conditions are sufficient for general identification and are necessary and sufficient for exact identification. Second, we show that these conditions can be easily implemented and that they apply to a wide class of identifying restrictions, including linear and certain nonlinear restrictions. Third, we show that the rank condition for exactly identified models amounts to a straightforward counting exercise. Fourth, we develop efficient algorithms for small-sample estimation and inference, especially for SVARs with nonlinear restrictions.

We study how a fiscal expansion via infrastructure investment influences the dynamic impacts of monetary stimulus on credit allocation. We develop a two‐stage approach and apply it to the Chinese economy with a confidential loan‐level data set that covers all sectors. We find that infrastructure investment significantly weakened monetary policy's transmission to credit allocated to private firms, while reinforcing the monetary effects on loans to state‐owned firms. This fiscal‐monetary interaction channel is key to understanding the preferential credit access enjoyed by state‐owned firms during the stimulus period. Consequently, monetary stimulus crowded out private investment and decreased capital allocation efficiency.