search

ankargren / mfbvar

R package for Mixed-Frequency Bayesian VARs
https://ankargren.github.io/mfbvar
38 stars 21 forks source link

n_reps missing when using steady-state prior #15

Closed dcuoliveira closed 3 years ago

dcuoliveira commented 3 years ago

Im getting an issue of absence of "n_reps" variable when estimting the model, although i have it as an input in my prior.

Below you can see the code i tried:

rm(list=ls()) library('mfbvar') data("mf_usa")

mfbvar_prior = set_prior(Y = mf_usa, n_lags = 4, n_reps = 1000)

prior_intervals = matrix(c(1, 3, 4, 8, 1, 3), ncol = 2, byrow = TRUE) moments = interval_to_moments(prior_intervals) mfbvar_prior = update_prior(mfbvar_prior, d = "intercept", prior_psi_mean = moments$prior_psi_mean, prior_psi_Omega = moments$prior_psi_Omega) plot(mfbvar_prior) mfbvar_prior = update_prior(mfbvar_prior, n_fcst = 24) summary(mfbvar_prior)

mod_ss_iw = estimate_mfbvar(pmfbvar_prior = mfbvar_prior, prior = "ss", variance = "iw")

and the error:

Error in mget(names(formals())[names(formals()) != "..."], sys.frame(sys.nframe())) : argument "n_reps" is missing

ankargren commented 3 years ago

That is because you have typo:

mod_ss_iw = estimate_mfbvar(pmfbvar_prior = mfbvar_prior,

should be

mod_ss_iw = estimate_mfbvar(mfbvar_prior = mfbvar_prior,