Quote from: Deprived on June 30, 2013, 08:52:38 PM

Quote from: crypto-finance on June 30, 2013, 07:15:00 PM

I wrote a first draft of an article explaining what I believe to be an arbitrage situation with DMS.SELLING and mining difficulty futures on icbit.se (hence my previous questions about the cashflow of DMS.SELLING).

http://www.crypto-finance.com/difficulty_arbitrage.pdf

I'm hoping that financially/mathematically skilled people can let me know about any mistakes. I hope you guys enjoy the read.

One problem is that it will remain a theoretical arbitrage rather than a practical one. For it to be useful in practice you'd have to rely on the market correctly pricing SELLING at the settlement on icbit.se.

A second issue is that if you assume the icbit.se projected difficulty rise is correct then you could almost certainly make significantly more money by just buying PURCHASE, splitting it and selling MINING - as their projection is a LOT worse than the one the market appears to be using in its current valuation. I haven't, however, looked at icbit - and it may well be that their projection is just what they're using when selling futures - and so is nowhere near what even they believe (but rather a projection sufficiently high that they're confident of profit). That's in the same way that if you go to a bookie and they offer you an even money bet they do NOT believe you'll win half the time.

Regarding the first issue: that is totally correct. I mention it in the conclusion of this paper. It makes it more of a "statistical arbitrage" than a strictly risk-free one. If the market is irrational about it`s valuation of DMS.SELLING, then there may be unexpected losses.

Regarding the second issue: that is the fact with hedging in general. The idea is that you sacrifice profit to reduce risk, or eliminate a particular source of risk (in this case, error in the estimation of difficulty on September 16th). By not hedging, I would essentially be making a bet that the futures contracts are correctly priced. I could also take the position that the futures contracts are wrong, and DMS.SELLING prices are correct, and instead short the futures... so why one over the other? I'm (personally) not willing to make any of those 2 bets, and would rather pay a little more to set up a hedge that "guarantees" (barring issue #1) profit, albeit a smaller one. Also, these futures contracts are quite cheap, so I don't really mind.

crypto-finance

Posts

Merits