SAS Econometrics software keeps improving its capabilities to meet increasingly complex analytical requirements. Among these improvements, the simulation smoother method is added to
the CSSM procedure so that you can obtain the joint sampling distribution of latent states and then generate any time series data by following the same pattern in the observed data or calculate forecasts and confidence bands for arbitrary functions of response variables. A new algorithm in the
CAUSALDISCOVERY procedure leads to a speedup of several orders of magnitude and enables the discovery of causal relationships among thousands of variables. The global optimization algorithm is now parallelized in the HMM procedure to make use of all available CPUs.
