A two regime Markov switching mannequin for the log of M2 velocity, 1959-2022, as advised by Rick Stryker:

And listed below are the regime possibilities, smoothed, filtered.

Right here’s the fitted values in opposition to precise, and the residual.

I received’t declare I’ve achieved any rigorous robustness exams, or that these outcomes are consultant. The information and software program can be found so one can consider how nicely a two regime Markov switching mannequin does for (log) velocity. For me, it’s not nice, most likely not significantly helpful for policymaking.

Apart from this specification being in logs, I don’t suppose these outcomes differ substantively from reader AS‘s.

If Steven Kopits meant reversion to the excessive velocity imply was imminent, nicely we haven’t seen it it. If he signifies that there’s a 3rd regime, nicely, we’re presently into a 3rd, lowest imply – not an intermediate imply – regime, in keeping with a 3 regime MS mannequin.