A mixed-frequency Bayesian vector autoregression with a steady-state prior

• We consider a Bayesian vector autoregressive (VAR) model allowing for an explicit priorspecication for the included variables' `steady states' (unconditional means) for data measuredat dierent frequencies. We propose a Gibbs sampler to sample from the posteriordistribution derived from a normal prior for the steady state and a normal-inverse-Wishart prior for the dynamics and error covariance. Moreover, we suggest a numerical algorithmfor computing the marginal data density that is useful for nding appropriate values for thenecessary hyperparameters. We evaluate the proposed model by applying it to a real-timedata set where we forecast Swedish GDP growth. The results indicate that the inclusionof high-frequency data improves the accuracy of low-frequency forecasts, in particular forshorter time horizons. The proposed model thus facilitates a simple and helpful way ofincorporating information about the long run through the steady-state prior as well asabout the near future through its ability to cope with mixed frequencies of the data.

Subject headings

SAMHÄLLSVETENSKAP  -- Ekonomi och näringsliv -- Nationalekonomi (hsv//swe)

SOCIAL SCIENCES  -- Economics and Business -- Economics (hsv//eng)

Publication and Content Type

vet (subject category)

rap (subject category)