Provides a quick way to visualize the lag decay
specification in a BVAR model for given parameters by computing the
variance of the prior VAR coefficients across various lags.
Usage
decay.spec(qm, p, lambda)
Arguments
qm
Periodicity parameter: either 4 or 12 for quarterly or
monthly data.
p
Number of lags
lambda
Lag decay parameter [>0], which is lambda3 in
the Sims-Zha BVAR specification in szbvar
Value
A time series of length p of the prior variances for each lag.
Details
Computes the relative decay in the prior variance of the VAR prior
across the lags from 1 to
p. Useful for visualizing the rate of decay or how tight the prior
becomes at higher order lags.
References
Sims, C.A. and Tao Zha. 1998. "Bayesian Methods for Dynamic
Multivariate Models." International Economic Review. 39(4):949-968.