Quote from: AnonyMint on September 05, 2014, 10:19:29 AM

Quote from: rpietila on September 05, 2014, 09:51:32 AM

Quote from: AnonyMint on September 05, 2014, 09:47:26 AM

Quote from: rpietila on September 04, 2014, 10:56:39 PM

Thank You Peter R! I will add this model to the toolkit when I try to explain to people why sometimes (2013) it rises 100x and sometimes (2012) hardly at all. Randomness.

Randomness is not the only possible explanation. Log-logistic adoption is another possibility for a declining rate of appreciation.

So it is not sufficient to say, "the model must be my fit, and randomness explains why the fit is off this year".

The burden of proof is on the "fit" that fits the data less (despite being more complicated), i.e. yours.

(I seriously think you are wasting your precious time, trying to argue with me about statistics. Tell me when I do likewise re: your strong point.)

Define 'more complicated' mathematically? Would it be what you are more familiar with?

Overfit is a problem if the curve fit or regression model is not correct.

I am trying to help both of us avoid confirmation bias error.

Even the concept of fit has a plurality of models.

How are you so certain that Bitcoin adoption has not slowed?? You can say it is just a lull due to randomness, but how can you be so confident of that? I am trying to be objective. I see evidence it is slowing. It might be randomness, but it also might not be. I also see much confirmation that Bitcoin is not being adopted by most of those who hear about it, because there are enormous barriers to adoption.

P.S. I nearly aced my Theory of Probability and Statistics course 30 years ago. But correct I don't use it in practice. But that distinction seems to be irrelevant in this case, because we have to choose the correct model before we can apply statistical discipline. Or we need some very higher order statistical methods to try to determine which model is correct.

It is simply impossible to "choose the right model" with certainty. It is impossible to say today if bitcoin adoption has slowed or not.
- Statistically, the past is the best prediction of the future : it would mean adoption has not slowed and will catch up through sudden big moves of mass adoption and we will follow an exponential trendline which is the best model for the past so far.
- Empirically, adoption in 2014 has been disappointing which could mean we are invalidating the current model, yet there are preparations being made that would allow a greater wave of adoption.