I have used Stata and GRETL for univariate and multivariate time series analyses. And today decided to share my opinion with the community. I have concluded that GRETL is a reliable software for time series analyses while Stata capabilities are still in development. An example, Stata can estimate Vector Error Correction Models, however, it can not provide estimates of the standard errors for the IRF's. Therefore, I can not recommend Stata fot this type of analysis whereas the opposite is true for GRETL.
In Stata, some time series estimators are written by users and then down the road, the company incorporates them in the core software. An example, rolling command for rolling regression. Other examples of this practice, cusum and prais commands. So, I would like to know some papers that compare the time series capabilities of Stata and GRETL in order to have a better understanding of the two softwares. Any help is appreciated. Thank you, Charles -- Charles Koss http://charlesonnet.blogspot.com