Released on 30/08/2018:
• | The marginal predictive moments for DSGE models can now produce back- and nowcasts provided that the model does not have any measurement errors. These predictions make sense when the observed variables have a ragged edge such that certain variables at the back- and nowcast horizon have missing values. |
• | Variance decompositions for the state variables have been added as a complement to the conditional variance decompositions. |
• | Added sequential estimation of the marginal likelihood for DSGE models. Model sequences can be selected and the tools run via items on the Model sequence menu-item on the File menu. |
• | Added the option of using an external DSGE model solver. |
• | Improved handling of error messages. |
Page url:
http://www.texlips.net/yada/help/index.html?4_40.htm