bvar |

Estimate a Bayesian VAR specification.

Syntax:

var_name.bvar(options) lag_pairs endog_list [@ exog_list]

bvar estimates an Baysian VAR. You must specify the order of the VAR (using one or more pairs of lag intervals), and then provide a list of series or groups to be used as endogenous variables. You may include exogenous variables such as trends and seasonal dummies in the VAR by including an “@-sign” followed by a list of series or groups. A constant is automatically added to the list of exogenous variables; to estimate a specification without a constant, you should use the option “noconst”.

Options

General options

noconst | Do not include a constant in exogenous regressors list. |

prior = keyword (default= “lit”) | Set the files as follows for prior types: “lit” (Litterman/Minnesota prior), “sznw” (Sims-Zha Normal-Wishart prior), “nw” (Normal-Wishart prior), “sznf” (Sims-Zha Normal-flat prior). |

initcov = keyword (default = “full”) | Set the (initial) residual variance-covariance: “uni” (Univariate AR estimate), “full” (full VAR estimate), “diag” (diagonal VAR estimate). By default, EViews uses the “initcov=uni” option so that diagonal elements of the prior residual variance-covariance can be obtained from the estimation of a set of univariate AR models. |

nodf | Degrees of freedom correction for initial residual covariance. |

l0 = arg | Set the residual covariance tightness hyper-parameter (for the Litterman prior; when the “prior=” option is set to the default “lit”). |

l1 = arg | Set the overall tightness hyper-parameter (for the Litterman prior; when the “prior=” option is set to the default “lit”). |

l2 = arg | Set the relative cross-variable weight hyper-parameter (for the Litterman prior; when the “prior=” option is set to the default “lit”). |

l3 = arg | Set the lag decay hyper-parameter (for the Litterman prior; when the “prior=” option is set to the default “lit”). |

mu1 = arg | Set the AR(1) coefficient dummies hyper-parameter (for the Litterman prior; when the “prior=” option is set to the default “lit”) |

mu5 = arg | Set the sum of coefficient dummies hyper-parameter (for the Litterman prior; when the “prior=” option is set to the default “lit”). |

mu6 = arg | Set the initial observation dummies hyper-parameter (for the Litterman prior; when the “prior=” option is set to the default “lit”). |

userpriors | Use user-specified priors as specified using the “usercoefs=”, “usercoefcov=”, “userhmat=”, and “userrescov=” options. |

usercoefs = name | Set the user-specified prior mean values for the Minnesota/Litterman (“lit”) and Normal-Wishart (“nw”) priors. It should be either a (coefficients per equation) by (endogenous variable) matrix or a vector. |

usercoefcov = name | Set the user-specified prior variance-covariance for the Minnesota/Litterman (“lit”) prior. It requires either a matrix or a vector. |

userhmat = name | Set user-specified diagonal elements of the prior precision matrix for the Sims-Zha Normal-Wishart (“sznw”) and Sims-Zha Normal-flat (“sznf”) options. A vector is required. A Minnesota-type of specification for the precision matrix is adapted and used here. In practice, the prior precision matrix is specified as a diagonal matrix |

userrescov = name | Set the user-specified diagonal elements of the prior scale matrix for the the Sims-Zha Normal-Wishart (“sznw”) option. A vector is required |

prompt | Force the dialog to appear from within a program. |

p | Print basic estimation results. |

Examples

var mvar.bvar 1 3 m1 gdp

declares and estimates an unrestricted VAR named MVAR with two endogenous variables (M1 and GDP), a constant and 3 lags (lags 1 through 3).

mvar.bvar(noconst) 1 3 ml gdp

estimates the same VAR, but with no constant.

var mvar.bvar(prior=nw) 1 3 m1 gdp

specifies the normal-Wishart prior.

var mvar.bvar(prior=nw, mu1=0.2, l1=0.2) 1 3 m1 gdp

specifies a normal-Wishart with hyper-prior values specified as mu1=0.2, lambda1=0.2.

vector(3) S0 = 1

vector(7) H0 = 1

var bvar.bvar(prior=sznw, userprior, userhmat = H0, userrescov=S0) 1 2 gdp inflation interest

declares and estimates a Bayesian VAR named BVAR with three endogenous variables (GDP, INFLATION and INTEREST), two lagged terms (lags 1 through 2) and a constant. The Sims-Zha Normal-Wishart (“prior=sznw”) prior is used with the user-specified parameter values for the diagonal elements of the coefficient precision (“userhmat=H0”) and the scale matrix (“userrescov=S0”) of the residuals.

Cross-references

See “Vector Autoregression and Error Correction Models” for details.