Interpretation of results about scicsr HOT 2 CLOSED

Hi,

• The stationary distribution describes the distribution of states "at equilibrium". In our experience we find that tends to match the observed distribution of states (in the data used to fit the model) quite well.
• The flux on the other hand measures transitions between states. In a system with good amount of signals of transition (see below) we find you can use these fluxes to estimate prospectively (ie at a future timepoint) the state distribution - this was what we did in the paper for the SARS-CoV-2 vaccination data (Fig 5).
• In our experimentation we did find occasions where we don't observe a good amount of signals for transitions - the flux plot would look very empty (you would only see one dot from the start to the end state, and nothing else). Looking at your first plot I would say this is the case.

If I were you I would also take a look at the flux plot for your other state - happy to take this discussion offline if you wish.

from scicsr.

I will close this issue, but if there are anything I can help, please open a new issue.

from scicsr.