Navigation:  Menu Commands > DSGE-VAR Menu >

 

Estimate Marginal Posterior Mode

 

Print this Topic Previous pageReturn to chapter overviewNext page

Run the marginal posterior mode estimation routine for the DSGE-VAR model. This routine maximizes the marginal likelihood function of the DSGE model parameter times the prior of these parameters. It can be determined by multiplying the likelihood function of the DSGE-VAR model with the prior of the VAR parameters conditional on the DSGE model parameters and then integrating out the VAR parameters; see, for instance, equation (A.2) in Del Negro and Schorfheide (2004).

The estimation routine can be performed for a selection of the hyperparameter λ that determines the weight on the prior for the DSGE-VAR model. The user can choose to skip estimating those selected λ values for which posterior mode results already exist on disk.

Apart from the marginal posterior mode estimate of the DSGE model parameters, the routine provide plug-in estimates of the VAR parameters. These plug-in estimates are obtained by maximizing the joint posterior of the DSGE-VAR where the solution of the VAR parameters is expressed as a function of the DSGE model parameters.

 

Additional Information

A detailed description about posterior mode estimation in the DSGE-VAR model can be found in Section 15.7 of the YADA Manual.

 

 


Page url: http://www.texlips.net/yada/help/index.html?estimate_marginal_posterior_mo.htm