O = BVAR.litterman(Rho,Mu,Lmb)
Rho [ numeric ] - White-noise priors (Rho = 0) or random-walk priors (Rho = 1), or something in between.
Mu [ numeric ] - Weight on dummy observations.
Lmb [ numeric ] - Exponential increase in weight depending on the lag; Lmb = 0 means all lags are weighted equally.
O [ bvarobj ] - BVAR object that can be passed into the VAR/estimate function.See the section explaining the weights on prior dummies, i.e. the input argument Mu.