Thanks for the link, I know about the model, but not in detail. Its main problem, from my point of view, is that there have been only two halving events, too few for more or less solid statistics.
The model has been working even BEFORE the first halving. So the number of halving is pretty irrelevant.
Model parameters considering 2009-2012 data points (before first halving) are very similar to ones obtained considering the whole BTC price history.