set_prior_cnw: sets up conjugate Normal-Wishart prior
In joergrieger/bvar: Estimation and forecasting of bayesian VAR models

Description Usage Arguments Details Value Author(s) References See Also

sets up conjugate Normal-Wishart prior

1
2
3

set_prior_cnw(mydata = NULL, factordata = NULL, no_factors = 0,
  coefprior = NULL, coefpriorvar = NULL, varprior = NULL,
  varpriordof = NULL, nolags = 1, intercept = TRUE)

`mydata`	a TxK xts-object needed for setting up the prior
`factordata`	for factor models additional time series for the factors are needed (not yet implemented)
`no_factors`	number of factors in a factor model (not yet implemented)
`coefprior`	double or () matrix with the prior for the VAR-coefficients. If only a scalar variable is provided the prior will be set to
`coefpriorvar`	double or () matrix with the prior on the variance of the VAR-coefficients. If only a scalar is provided the prior will be set to diag(1,K)
`varprior`	double or () matrix with the prior on Variance-Covariance matrix.
`varpriordof`	integer. The degree of freedom for prior on the Variance-Covariance matrix.
`nolags`	integer Number of lags in the VAR model
`intercept`	logical whether the VAR model has an intercept (TRUE) or not (FALSE)

This sets up the Normal-Wishart prior to use for bvar-estimation. The conjugate Normal-Wishart prior has the following form

α|Σ\sim N(\underline{α},Σ\otimes \underline{V})

and

Σ^{-1}\sim W(\underline{S}^{-1},\underline{v})

with \underline{α},\underline{V},\underline{v}, and \underline{S} the prior hyperparameters chosen by the user.

returns an S3-object of class "cnw"

Joerg Rieger

K. Rao Kadiyala and Sune Karlsson, Numerical Methods for Estimation and Inference in Bayesian VAR-Models, Journal of Applied Econometrics 12(2), 99-132

Gary Koop and Dimitris Korobilis (2010), Bayesian Multivariate Time Series Methods for Empirical Macroeconomics, Foundations and Trends in Econometrics 3(4), 267-358

Other priors: set_prior_minnesota, set_prior_ssvs, set_prior_uninformative

joergrieger/bvar documentation built on July 3, 2020, 5:34 p.m.