Arbitrage Strategy

[savagemp5]

Quote from Hammy27:

Oh it's not arbitrage in the sense that I buy the same stock on different exchanges to generate a profit. As you said, I wouldn't have an edge!
But I can do some stats on serial correlation, stationarity and so on. And cause I believe it's hard to compete on 2-stock basis I built a portfolio.

How I choosed the stocks? I asked Bloomberg to display me the ones I need to generate the maximum possible profit...
No seriously, I've got 2 criterias which I use for selection. I first choose a stock that I like and then I'll let Bloomberg find the 9 others I will need to build the strategy. So I do not fit anything. And: The stocks are the same for the whole time-frame.

But back to my initial question: if we assume that the backtesting is accurate, would you invest in such a strategy?
I'm currently not sure if I will... I will probably trade the portfolio on Bloomberg first for to see how it goes in the near future.

Don't explain. You're not arbing anything at all. Firstly you need capital, and secondly you have risk. If 1 day an airplane hits NYSE or if electricity fails on ALL US exchanges, that you cannot buy/sell on market, then its not an arb, then you have a risk. Arbs are suppose to be riskfree, and locks in profit instantly. eg. Box spread or synthetic split-strike.

Btw curve is very very easy. I think 15 year old can do it too. With good backtest data and tons of mathematics formula, you can easily find stocks that after long/short of a basket, or using relative value, make alpha.

Post your 3 year Sharpe ratio and max AUM on real time data before we talk. Try harder, Good luck

 

[xiaodre]

Disappointing.

I open up this thread and it's not arbing anything. Buddy, you gotta do something like look for differences in the price of gold on nymex versus the Aussie exchange, or the Jap exchanges, and then take the difference between the two in a trade. That's arbitrage trading.

 

[Shanb]

Quote from xiaodre:

Disappointing.

I open up this thread and it's not arbing anything. Buddy, you gotta do something like look for differences in the price of gold on nymex versus the Aussie exchange, or the Jap exchanges, and then take the difference between the two in a trade. That's arbitrage trading.

Pair trading is technically a form of arbitrage. It's called "Statistical Arbitrage" lol

To the OP, sounds like you are doing some basic pair trading? If I'm wrong let me know, but ya if you introduce some leverage into the equation you could have a good approach and make some solid money. There is of course some more exposure to tail risk when betting on convergence of spreads, add some leverage and things can get messy if your risk management isn't on point!

 

[savagemp5]

Quote from Shanb:

Pair trading is technically a form of arbitrage. It's called "Statistical Arbitrage" lol

To the OP, sounds like you are doing some basic pair trading? If I'm wrong let me know, but ya if you introduce some leverage into the equation you could have a good approach and make some solid money. There is of course some more exposure to tail risk when betting on convergence of spreads, add some leverage and things can get messy if your risk management isn't on point!

Pair trading is not stat arb.

 

[Shanb]

Quote from savagemp5:

Pair trading is not stat arb.

Pair trading is the most simplest form of Stat Arb. If pair trading is not stat arb then what is?

 

[newwurldmn]

Quote from Shanb:

Pair trading is the most simplest form of Stat Arb. If pair trading is not stat arb then what is?

If I buy SPY and sell NDX that's a pair trade, but not necessarily a stat arb strategy.

 

[savagemp5]

Quote from Shanb:

Pair trading is the most simplest form of Stat Arb. If pair trading is not stat arb then what is?

If you buy a pair, any pair that you think its called stat arb, then I'll ask you:

- If 1 day both products =$0, do you make or lose. If you lose $, its not an arb, and certainly not statistically at all and its a risk.

- If 1 day a major news hit 1 of the pair, does the other product move together ? If it doesn't again its not an arb, and there's risk involve.

Don't mixed up pair trading, hedging, delta one, Neutral, Spread trading with stat arb.

 

[bone]

I have done consulting work for two very prominent HFT electronic trading firms in Chicago and Manhattan [ household names in the biz you would instantly recongnize ] - they both spread trade equity pairs and baskets on an automated basis at very high frequencies, and they both call it 'stat arb'.

One of the firms does so much volume, it actually owns a brokerage named after a citrus fruit.

The other firm does so much volume they really should have their own brokerage - if there was any money in it to be had.

Taking little nibbles out of highly correlated spread differentials on a milliseconds timeframe. All day long.

 

[ElectricSavant]

I always called pair trading correlation trading...But what do
I know?...I trade Forex...

ES

Quote from Shanb:

Pair trading is the most simplest form of Stat Arb. If pair trading is not stat arb then what is?

 

[Shanb]

Quote from savagemp5:

If you buy a pair, any pair that you think its called stat arb, then I'll ask you:

- If 1 day both products =$0, do you make or lose. If you lose $, its not an arb, and certainly not statistically at all and its a risk.

- If 1 day a major news hit 1 of the pair, does the other product move together ? If it doesn't again its not an arb, and there's risk involve.

Don't mixed up pair trading, hedging, delta one, Neutral, Spread trading with stat arb. [/QUOTE

Betting on the convergence of a Cointegrating spread is stat arb. You are betting on mean reversion because of the statistical property of the time series.

A fundamental or news event can of course come out and eliminate the cointegrating relationship between two assets. Its upto the trader to know when that has happened! Cointegration does not exist into perpetuity! When you are betting on the convergence of a spread you are betting that the cointegrating relationship will exist going forward.