The Conventional Impulse Response Prior in VAR Models with Sign Restrictions

Some studies have expressed concern that the Gaussian-inverse Wishart-Haar prior typically employed in estimating sign-identified VAR models may be unintentionally informative about the implied prior for the structural impulse responses. This paper discusses how this prior may be reported and makes explicit what impulse response priors a number of recently published studies specified, allowing the readers to decide whether they are comfortable with this prior.