The DSGE team at the Federal Reserve Bank of New York is pleased to introduce a forecasting component to the DSGE.jl package.
Our Liberty Street Economics blog post gives a high-level overview of the new functionality. In this technical post, we provide more details about our performance improvements, the principles that guided our design, and how we took advantage of parallel programming with Julia. We hope this perspective will be helpful for economists and other social scientists using Julia.
We also are pleased to introduce StateSpaceRoutines.jl, which implements common state-space algorithms. We provide a Kalman filter as well as two Kalman smoothing routines and two simulation smoothers. All routines can accommodate time-varying state-space matrices. We believe that these implementations may be the first time-varying state-space routines available in a Julia package, and hope that they can be of use to the community.
As always, we are happy to hear your comments and suggestions!