Choose the Hyperparameters Set of SSVS-VAR using a Default Semiautomatic Approach

This function chooses \((\tau_{0i}, \tau_{1i})\) and \((\kappa_{0i}, \kappa_{1i})\) using a default semiautomatic approach.

Usage

choose_ssvs(
  y,
  ord,
  type = c("VAR", "VHAR"),
  param = c(0.1, 10),
  include_mean = TRUE,
  gamma_param = c(0.01, 0.01),
  mean_non = 0,
  sd_non = 0.1
)

Arguments

y

Time series data of which columns indicate the variables.

ord

Order for VAR or VHAR.

type

Model type (Default: VAR or VHAR).

param

Preselected constants \(c_0 << c_1\). By default, 0.1 and 10 (See Details).

include_mean

Add constant term (Default: TRUE) or not (FALSE).

gamma_param

Parameters (shape, rate) for Gamma distribution. This is for the output.

mean_non

Prior mean of unrestricted coefficients. This is for the output.

sd_non

Standard deviance of unrestricted coefficients. This is for the output.

Value

ssvsinput object

Details

Instead of using subjective values of \((\tau_{0i}, \tau_{1i})\), we can use $$\tau_{ki} = c_k \hat{VAR(OLS)}$$ It must be \(c_0 << c_1\).

In case of \((\omega_{0ij}, \omega_{1ij})\), $$\omega_{kij} = c_k = \hat{VAR(OLS)}$$ similarly.

References

George, E. I., & McCulloch, R. E. (1993). Variable Selection via Gibbs Sampling. Journal of the American Statistical Association, 88(423), 881-889.

George, E. I., Sun, D., & Ni, S. (2008). Bayesian stochastic search for VAR model restrictions. Journal of Econometrics, 142(1), 553-580.

Koop, G., & Korobilis, D. (2009). Bayesian Multivariate Time Series Methods for Empirical Macroeconomics. Foundations and Trends® in Econometrics, 3(4), 267-358.