Don’t worry about the buzzwords in finance. Statistical arbitrage is basically pairs trading. Pairs trading, in its simplest form is buying or selling a pair of stocks based on its relationship to each other.
Often, two stocks in the same business domain are ‘synched’ – meaning that their price patterns follow each other. Sometimes they come out of the sync and that is when a statistical arbitrage opportunity appears to make some serious profits.
The good thing about pairs trading is, it is market neutral, meaning that it is not depended on the market move but the relationship of the two paired stocks only. This means that you could potentially make good profits whatever the market conditions are at a given time.
The relationship between the two stocks can be described by the ‘correlation coefficient’ which is basically the strength of relationship between a dependant variable and an Independend variable.
- Perfect negative correlation (-1) exists when the two securities move in opposite directions (i.e., stock A moves up while stock B moves down);
- Perfect positive correlation (+1) exists if the two securities move in perfect unison (i.e., stock A and stock B move up and down at the same time); and
- No correlation (0) exists if the price movements are completely random (stock A and stock B go up and down randomly).
An example can be given as follows (copyright of Pairtrade Finder Software):
In this example you see the price relationship between KO and PEP and the areas of interest for pair trading.
Pepsi (PEP) and Coca Cola (KO) are different companies that create a similar product, soda pop. Historically, the two companies have shared similar dips and highs, depending on the soda pop market. If the price of Coca Cola were to go up a significant amount while Pepsi stayed the same, a pairs trader would buy Pepsi stock and sell Coca Cola stock, assuming that the two companies would later return to their historical balance point. If the price of Pepsi rose to close that gap in price, the trader would make money on the Pepsi stock, while if the price of Coca Cola fell, he would make money on having shorted the Coca Cola stock.
The dog and the owner
Murray (1994 – American Scientific) has given an excellent example in his research paper regarding pairs trading.
Suppose you see two drunk people (i.e., two random walks) wandering around. The drunks don’t know each other (they’re independent), so there’s no meaningful relationship between their paths.
But suppose instead you have a drunk walking with his/her dog. This time there is a connection. What’s the nature of this connection? Notice that although each path individually is still an unpredictable random walk, given the location of one of the drunk or dog, we have a pretty good idea of where the other is; that is, the distance between the two is fairly predictable. (For example, if the dog wanders too far away from his owner, she’ll tend to move in his direction to avoid losing him, so the two stay close together despite a tendency to wander around on their own.) We describe this relationship by saying that the drunk and her dog form a cointegrating pair and we can scientifically calculate the correlation coefficient.
OK I got it! Show me how can I start using this technique?
All good in theory so how do we use in practice. Well I am working on this and paper trading on number of stocks for about 6 month or so. The results are very encouraging. I am planning to start live trades on pair-trading very soon and I will report those on Twitter just like my day trading trades. This will not replace day trading at all, just would like to add another tool in my trading toolset.
There is no need to re-invent the wheel. There are number of software programmes and subscription based services on the Internet where you can download and easily find pairs. They work quite well. I will not recommend any services as they are all different and has positives and negatives. Try and see what works for you best.
Or like me, you can start from scratch and create your own pair finder software programme. I have created the software using matlab (I realise this is not for everyone, although it is not very difficult it needs certain technical knowledge especially in matlab programming – there are a lot of free to use libraries out there if you wanted to do it).
Please note, pair trading is not risk free. Exactly the opposite in fact, it can be very risky if you choose the wrong pairs. It can be also time consuming (you will probably need a separate account and funds to try this ), it might takes from days to weeks for the profitable setups to take place.
Overall, I am very encouraged and will continue to test the system a little while before starting my live trades soon. I will update this blog again with the live results soon.