He's right. Without shorting, options, and future contracts... it becomes impossible for BTC to stabilize in the wake of media exposure. Add on to the fact that the majority of BTC users seem to be idiots (ie: they look at the price as some sort of indicator of BTC penetration, as opposed to more useful statistics), and you've definitely got a situation where bubbles will continuously form.
Anyway, I don't necessarily think he's right. There will always be some function that fits some data... and he may have gotten lucky this time that data fits his model. Either way, it is certainly an interesting piece to read. And his model seems to have solid theory behind it.
I agree that a futures market would be a stabilizing force, but other volatile commodities are still highly volatile even with futures markets.
The reason most currencies aren't volatile is that they have a central bank behind them actively manipulating their supply to make sure they are stable relative to some other asset or basket of goods. In the case of the USD, the dollar is roughly pegged to CPI. Unbacked commodities that aren't pegged by a central authority tend to be on a highly volatile random walk, with or without futures markets.
EDIT: Looks like I mistyped, maybe it looked like I was saying the opposite.
Its all comparative. :-p It will still be more volatile than currencies, but a hell-of-a-lot better than it is right now.
The speculators aren't in any danger of disappearing, no sir, this is just an opportunity to buy 'cheap' bitcoin and increase their holdings. Because obviously squirrelling away as much of it as possible will help it become a viable currency...
We now have a testable prediction. Let see if bitcoin actually falls to to 20 and 10 dollars per bitcoin within 4 weeks.
Not necessarily as a stable value, but in the last few days it's bounced around madly between about 50 and 100 dollars. I think at this point it's fair to say that nobody has any idea of the real 'value' of a bitcoin, and mad speculation is still the order of the day.
Nonsense. They have no use, not even as a status symbol or as a practical unit of exchange. The “real value of a bitcoin” is zero, and the fact that this isn’t obvious to everyone is, frankly, astonishing.
Traditionally, the idea with a bubble is that everyone (well, almost everyone) knows it's a bubble, but no one seems to know when it will pop or how far it will fall.
Would this same model have fit the 2008 stock market collapse? Would it have accurately showed when and where the bottom was?
Would this same model have fit the BTC curve as well if the dataset had started 100 or 200 days earlier or later?
Just some curiousity about a model I'm hearing of for the first time.
The bubble was specifically in Credit Default Swaps, a derivative of the bond market. The problem here is that CDSes were untracked and unregulated. No one knew there was a bubble because there was no way to see the "fair price" of a CDS. Companies were making deals on CDSes in their backrooms, away from exchanges.
When all of the companies involved in CDSes failed (because people failed to pay their subprime mortgage loans), it killed the banking industry... even those unrelated to the bubble. When your business partner goes bankrupt, you're also in danger. Again: there were lots of factories who couldn't get a loan to pay their workers... because the bank they relied on died in the whole crisis.
This leads to factory closings, lots of people losing their job, and then a general Stock market crash.
But again, Stocks weren't the bubble in 2008. The Credit Default Swaps in the bond market was the problem.
You'd do better to apply it to a 2001 tech index.
The difference is a bubble is driven by greed, and "greater fool" behaviors turning to fear and panic selling. De-levering contagion is driven by a position going down triggering margin calls which necessitate selling other positions which drive down prices which furthers the cycle.
https://en.wikipedia.org/wiki/Random_walk_hypothesis
https://en.wikipedia.org/wiki/Nash_equilibrium
https://en.wikipedia.org/wiki/Efficient-market_hypothesis
I though you guys were more smart than this.
And also a new funded startup coming: http://siliconangle.com/blog/2013/04/11/coinsetter-the-newes...