Arbitrage Strategy

[syswizard]

Quote from Rationalize:

If you're spread trading small size and turning it over many many times per day, then there is somewhat less risk than an LTCM blow-up-the-word scenario. Many small trades with positive expectancy. Similar idea, different risk.

Exactly. Their positions required a longer term holding committment ....and that's what did them in.

 

[Rationalize]

Quote from syswizard:

Exactly. Their positions required a longer term holding committment ....and that's what did them in.

I think there was a bit more to it than that, and as I understand it, the portfolio on net unwound profitably over time. A lot of time.

 

[Hammy27]

ok... now the thread is about LTCM

Since I like to talk about stat-arb I post a simple graph showing a pair-strategy.
If you entered the strategy when the spread is below 0.06 (exit@0) and shorted the spread when it's above 0.04 (exit@0), then you would have earned 12'808 CHF (or a multiple of). Since the strategy is $-neutral your initial amount is 0CHF (I invested 20k CHF on each side). The strategy would have generated 7 Trades in a 3 years timeframe with an average holding period about 36 days. All the trades are profitable.

If I run the spread through an Augmented Dickey Fuller Test I get a ADF stat of -3.68 with a p-value about 0.024. That means that with a probability of 97.6% the spread is cointegrating.