Compute the sample or population-based conditional correlations for a DSGE-VAR at initial or marginal/joint/DSGE posterior mode parameter values, or a sample from the prior or the posterior distribution of the parameters. The conditional covariance matrix of the observed variables is calculated through the DSGE model by assuming that all shocks but one is zero.
The population-based correlations are quickly computed since their values are known once the parameters are treated as know. The sample-based correlations, on the other hand, need to be estimated through simulation. The number of simulation paths per parameter value is determined in the forecasting frame on the Miscellaneous tab, while the number of parameters is determined by the number of post burn-in posterior draws and the percentage use of posterior draws for impulse responses, variance decompositions and other functions of the parameters. These latter values are set in the posterior sampling frame on the Options tab.
Additional Information
• | A more detailed description about the conditional correlations for the DSGE-VAR can also be found in Section 16.6 of the YADA Manual. |
Page url: http://www.texlips.net/yada/help/index.html?conditional_correlations2.htm