Code

JFerence Toolbox

The JFerence toolbox is a suite of MATLAB codes used for conducting joint inference about impulse response functions or forecasts in Bayesian time series models. It computes error bands using Bonferroni, Šidák, sup-t or min-max estimators, incorporating calibration steps to ensure the nominal probability content is achieved. Before constructing the error bands, a model must be estimated using any Bayesian method, and posterior draws of the relevant quantities are used as inputs of the JFerence routines.

Further information about the joint inference methods implemented in the toolbox is avilable in my job market paper.

The latest version of the toolbox is available on GitHub: JFerence Toolbox 1.0.