A video discussing what "steady state" means in the context of Stock-Flow Consistent models.
It uses a tool that allows users to examine the time series created by the sfc_models package. It's not yet incorporated into sfc_models; the code used in the video currently will not run unless you install the latest source code from GitHub. The GUI is in the brianr747/sfc_gui repository.
One thing I did not remark on was the effect on statistical analysis. Most of the time, the system was in steady state, and we could not draw much information from the time series data. All the useful information was embedded in the short transition from one steady state to another. (In engineering, we use things like the rise time and overshoot to get an approximation of how the system operates.) That is, we cannot conclude anything about the statistical usefulness of data solely based on the number of data points; we also need to take into account the operating characteristics of the system. In other words, we cannot do empirical analysis without some theory behind it.
(c) Brian Romanchuk 2017
No comments:
Post a Comment
Note: Posts are manually moderated, with a varying delay. Some disappear.
The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.
Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.