Navigation:  Menu Commands > Learning Menu > View >

 

Posterior Mode Results

 

Print this Topic Previous pageReturn to chapter overviewNext page

View all the posterior mode results for the DSGE model under adaptive learning. The results are divided into the following categories:

Model information: The data includes the files used by the estimation routine.
Sample and data information: The data contains the selected sample, the length of the training sample, the estimation sample, names of variables, etc.
Optimization information: The selected optimization routine, the optimization time, the number of iterations, the tolerance level used, and the value of the convergence function. For csminwel, the results also include detailed output about each step.
Posterior density results: Gives you data on the different terms of the log posterior at the initial values and at the posterior mode values. The values of the log posterior over all the iterations are also provided, along with the value of the convergence function and the maximum absolute value of the numerical gradient of the log posterior.
Parameters: Offers data regarding the transformed and the original parameters. The data includes the mode of the joint posterior, the initial value, the standard deviation as given by the inverse Hessian at the posterior mode, prior distribution information. Morever, parameters that are functions of the estimated parameters are shown at the posterior mode and at the initial value. Parameters with a maximum absolute correlation (using the inverse Hessian at the mode) greater than or equal to 0.7 are also listed.
Forecast errors: Presents the 1-step ahead mean and mean squared forecast error using the initial parameter values and the posterior mode values.

 

 


Page url: http://www.texlips.net/yada/help/index.html?posterior_mode_results4.htm