Piggybacking on yesterday’s post, here is a follow up by another anonymous quant, Nick. The title of the article is a little misleading, but makes sense given his method for scientific discover, Falsifiability:
According to the great modern scientific philosopher Karl Popper, testing a hypothesis for an incorrect outcome is the only reliable way to add weight to the argument that it is correct¹. If rigorous and repeated tests cannot show that a hypothesis is incorrect, then with each test the hypothesis assumes a higher likelihood of being correct. This concept is called Falsifiability.
The author attempts to falsify PlanB’s stock-to-flow model, but all of his statistical tests “fail to reject the hypothesis that stock-to-flow is an important non-spurious predictor for the value of Bitcoin.” Thus, his battery of additional statistical testing are unable to falisfy the stock-to-flow model, which in turn adds weight to the validity of the model (and its predictive ability).
Interestingly, Nick is able to find a statistically significant cointegration equation between (log) stock-to-flow and (log) market value, which implies that the relationship between the variables is super consistent. Or with the metaphor Nick uses:
To understand cointegration we give a simplified example of a drunk and her dog. Imagine a drunk walking her dog home on a leash. The drunk is walking all over the place, unpredictably. The dog walks pretty randomly as well: sniffing trees, barking, chasing scratching — just generally being a mutt. However, the dog’s overall direction will be within the length of the leash of the drunk. We could estimate that for any point on the drunks walk home, the dog will be within leash length of the drunk (sure it might be on one side or another, but the dog will be within leash length). This bad simplification is a rough metaphor of cointegration — the dog and the owner are moving together.
Cointegration of the variables is even more noteworthy than the 95% correlation between variables as determined by PlanB, i.e. the correlation of stock-to-flow and market value could be very high, but be wildly inaccurate or not really mean anything. However, with cointegration of the variables, we can be sure that the correlation is in fact meaningful and the stock-to-flow and market value move together, like the drunk and his dog via the leash.
In the end, Nick comes up with an even more apt metaphor:
To illustrate this with a metaphor: if we consider the value of bitcoin as the drunk then the stock-to-flow is not really the dog she walks — it is more like the road she walks on. The drunk will wander all over the road, sometimes stopping, slipping up and missing a turn here and there or even taking short cuts along the way; but generally, she will follow the road home.
In short, Bitcoin is the drunk and Stock-to-Flow is the road home.
That sounds about right considering bitcoin’s (drunken?) price volatility. But it also sounds really good to me considering the road the stock-to-flow model predicts the price to follow in the near future:
One thought on “Falsifying Stock-to-Flow as a Model of Bitcoin Value”