Dear scholars, I Bayesian SVAR better than Recursive SVAR when using annual data considering the low frequency. Recommended references will be appreciated.
First of all, you need to understand that, Bayesian works better on small samples. Regardless the sample size, the Markov Chain stimulate the sample into larger samples, let's say 20,000!
In Bayesian, the prior/believe about a certain variables are very important to consider. You assign a prior to every parameters you want to estimate. However, the choice of prior is also important! Informative prior, Jeffrey prior, or formative prior!