Pairs Trading

[Trader333]

anley,

I disagree almost totally with what you have said. I am using pair trading on Nasdaq stocks and it has proven to me to be the lowest risk to reward strategy I have ever used.

But taking some of your specific comments:

And then once you factor in the cost of doing business (not just comms but 2 bid/offer spreads) and profitable pairs trading can hinge on transaction costs especially if you short term trade.

The bid offer spread frequently works in my favour as I buy on the bid and sell on the ask effectively capturing the spread which I do when I consider the spread to be too great which to me is anything over 2c. There are members of T2W who have seen me do this in realtime so know that it is true. When trading 500 shares for each stock in the pair trade they only need to move 2c for me to have covered my transaction costs. Also knowledge of how to route orders is of critical importance and I have an edge over most people in doing that as well.

Yesterday afternoon I pair traded for a net profit of 60 points and a friend of mine pair traded the evening for a profit of 78 points.

For both of these trades the market didnt go against the position by more than 5 points and the reward was 12 times that for one trade and 15 times that for the second.

Almost everyday we are able to identify stocks to pair trade and the rewards are always way above the risk and any commissions.

I dont exclusively pair trade but when I do I have estimated that the chance of loss is reduced by around 75% when compared to straight directional plays. This is based on results that I have been monitoring and will continue to monitor.

That said I agree with Grey1 in that it is not easy to determine which stocks to pair trade. I and others have put in an enormous amount of work to get to the point where we are now and there is still more to be done in refining the strategy. But it does work and I am confident that it will continue to do so in the future but I would never try it on UK stocks only the Nasdaq.



Paul

 

[Rognvald]

May I suggest that this thread which has thrown up the interesting topic of pair trading be divided to highlight the pair trading issue.
May I also request that Sharkey/Moderators give some thought and energy to extracting/linking thread content and/or grouping it by subject with some cross referencing. There is a wealth of information on this site but the gems are often well hidden & are sometimes quite difficult to access. Those who use the site frequently are better placed in this respect than the more casual user.
The site is growing very fast and if some sort of subject based archiving does not take place soon the sifting task will become too large and the valuable material will largely go to waste.
Could not an effort be made to canvass individuals to take on this work as a project?

 

[sandpiper]

Grey1,

"What did I have in my head". Some crap relating to trading negatively correlated sector stocks. As you've said, nowt to do with pairs trading. If anything, more to do with diversification, but nevertheless irrelevant to the current discussion. .

 

[BBB]

Wow!

I know a fair bit about spreads (futures) and options (maybe a bit rusty now!)

I assumed pairs trading was the same as a futures spread. With the BT/VOD example, it would seem this is not the case or I am obviously wrong.

PLEASE, PLEASE could someone direct me to a site, book or previous thread where I can learn more about Pairs.

 

[Trader333]

BBB,

Read all of Grey1 posts, he has done more to bring this knowledge to T2W than anyone else and is the resident expert in my view.


Paul

 

[anley]

Of course there are always exceptions to the rules but to me the jist of this whole thread is that somehow pairs trading is easier, more profitable and less volatile than standard directional trading.

I'm saying that the argument is flawed because with pairs trading you're still trading a markets direction.

Paul

I don't want an argument and am not looking for one but how when you obviously know what you're doing in the markets make a statement like this;

"if I am long one stock and short another then news will cause a ramp up or ramp down of both but my positions effectively cancel each other out in this case so why would I get hammered ?"

What happens if you're long share ABC and short XYZ and then something major happens in the market. It's very possible for ABC to collapse while XYZ stays firm. In that situation you'd be looking at some major losses. The profit on one would not take care of the loss on the other.

The other problem with pairs trading is that the relationships are fickle and often only make themselves apparent after the fact, so when you get ready to trade them the goal posts move without you knowing.

I saw a classic example of this in the FTSE/DAX spread back in about 1997-1998. A few people in London made a lot of money trading this spread for about 6 months, in and out, in and out. But then as I said above the relationship that had been so apparent and strong suddenly changed, only the market didn't have the courtesy to inform anyone.

So you know what happened, a lot of money was now given back to the market with some horrendous losses for a few of the bigger players. I know this first hand because I was one of the main brokers doing all the business.

So again my main point is that Pairs trading gives the illusion that's it's easier and possibly more profitable but in reality it's not, it's just the same as directional punting. Having said that some are no doubt more suited to it than others, just the same with spread trading in futures. Normally in futures you're a spreader or a directional player and very few people do both well.

I would say that the average private client starting out trading pairs is someone who hasn't got on that well with directional trading. So he says to himself, oh here's a way to take the overall direction out of the market, and getting the overall direction of the market wrong was why I lost money or didn't make money in the past. But then as I've said before he soon finds out that yes the ftse direction doesn't matter too much anymore but then the future direction of Tech/Gold or Astra/Glaxo does. And here's the kicker, if you chart those two relationships the chart looks like just another share, index, currency etc. It trends, it has ranges it has quiet days and busy days, just like the overall direction of the stockmarket.

So the question now becomes, do you go long, short or flat the spread. And this is just the same question that someone looking to trade a stock outright is faced with. It's all about numbers folks, is the number going up or down, FTSE or BT/VOD pair?

 

[Trader333]

anley,

As you say there is no point in arguing over this so I wont post after this because you clearly have strong views on this that I just dont agree with.

Again taking your quote:

What happens if you're long share ABC and short XYZ and then something major happens in the market. It's very possible for ABC to collapse while XYZ stays firm. In that situation you'd be looking at some major losses. The profit on one would not take care of the loss on the other.

Of course this can happen but success in trading is about probabilities and the above mentioned event you would lose just as badly with a directional play as a pair. As I have already said my aim is to reduce risk and that is what pairs trading has enabled me to do, not eliminate it just reduce it.

As for saying that the relationships are fickle well I am happy to be making consistent profits day in and day out with these fickle stocks, I just must be unbelieveably lucky then because it keeps increasing my account.

When I trade them the goalposts dont move I know where I am aiming for and I am out of the trade at specific points unless one of the stocks just keeps making more and more money in which case I trail a stop until I get taken out.

Your comment that:

So again my main point is that Pairs trading gives the illusion that's it's easier and possibly more profitable but in reality it's not, it's just the same as directional punting

I couldnt disagree more with this. If you know and understand how to pair trade it can and is very profitable with a hugely reduced risk.

I know you wont agree and that is fine, I have nothing to prove and nothing to sell and only come on these boards to help others where I am able to. We will just have to disagree and go our separate paths in trading.



Paul

 

[BBB]

Anley - How the devil do you go short a spread?

Quote: So the question now becomes, do you go long, short or flat the spread. And this is just the same question that someone looking to trade a stock outright is faced with. It's all about numbers folks, is the number going up or down, FTSE or BT/VOD pair?

Sorry if I'm being pedantic! I know loads of people who trade both spreads and 'single sided' trades very well. It's all about flexibility and your own current circumstances. A spread may be an excellent idea if you're off on holiday and able to monitor the position every other day or so, while continue day trading on your return! This way you earn money while lying on a beach surrounded by chicks and boozing it up!

 

[Grey1]

Anley ,

Quote " What happens if you're long share ABC and short XYZ and then something major happens in the market. It's very possible for ABC to collapse while XYZ stays firm. In that situation you'd be looking at some major losses. The profit on one would not take care of the loss on the other.


Answer:-- of course this will happen if you are not doing it right..The whole idea is to select correct stocks hedged against each other.. It make all the difference if you know how to pair trade.
Have you asked yourself why stock XYZ stayed firm while Stock ABX collapsed.. perhaps you should ave gone short stock ABC at the outer VWAP ..

Quote "The other problem with pairs trading is that the relationships are fickle and often only make themselves apparent after the fact, so when you get ready to trade them the goal posts move without you knowing. "


Not correct at all.. You watch stocks all day to develope into a good pair trades before you trade them ... In another word you let the market to confirm your finding .. No amount of analysis is of any value if market does not feel the hedged stocks are suitable.. I have posted few real time pair trades if you care to look for them ..

Quote " I saw a classic example of this in the FTSE/DAX spread back in about 1997-1998. A few people in London made a lot of money trading this spread for about 6 months, in and out, in and out. But then as I said above the relationship that had been so apparent and strong suddenly changed, only the market didn't have the courtesy to inform anyone. "

Antley , the fact that stocks or indecies go opposite ways does not qualify them as a pair trade..? On top of that there must be a criteria for the end of spread widening other wise the spread gets wider and wider and the trader will get hammered .. if it was as simple as what you have posted then why donot we hedge two stocks from NAZ one which is 5% down and the other 5% up and whohhoo we have 10 % spread .. we hedge one against the other and we are in $$$..


How long does it take to become a dentist? Why should trading be any different ...

Regards..

 

[Grey1]

I have been coaching 4 of my dear freinds from this BB on various advanced subjects of trading and their trading performance has been excellent. .. Of course it is not easily possible to transfer expertise using PM 's and emails but the good news is that they have become profitable using these two means of communication.

One of the techniques has been the basic understanding of VWAP and its application on pair trading ..

PS:- Before I am being accused , NO I have not or ever will charge a fellow trader for passing knowledge.. I have the market to ripp off and not my fellow traders.. these guys do understand the concept of statistical edge and do not expect Miracles . neither do I ..

regards

 

[one/two]

Grey 1

your helping more then then just 4 members of this this BB

keep up the good work.

P.S how do we go about changing the title the the thread, it is almost certain to be lost if left as it is.

 

[Grey1]

one/two

I used to run a private forum for advanced Traders few years back. On that BB we used to also have a read only forum which acted like a library of trading strategies and so on ..

All the useful info and trading techniques was then easily accessible by all members and would never get lost in a cloud of personal abuse as you can currently see on the "follow my trade " thread.. There all excellent contributors on this BB such as NAZ,Chinawhite,Citytrader,Trader333,, too many to mention .. it is simply a shame all the knowedge gets overriden by sideline discussion which make no one a single penny ..

I think the name of the this thread should be changed.

regards

 

[one/two]

Grey 1

is it possible to access the Forum anymore
it would certainly make an interesting read.

 

[Grey1]

The forum was closed because there was a problem with Realtick and their rep in UK . The forum was part of www.ird.com


regards

 

[ChartMan]

I've changed the title. If you would prefer a different title, let me know.
ChartMan.

 

[Grey1]

Just a few points on pair trading ..

Pair trading or in general market neutral strategies are meant to reduce risk , other wise there is no point doing it.

In general there are four ways of hedging instruments against each other..

1) Pick two positively correlated instruments and hedge them against each other.. This needs an extensive correlation analysis and in my view is kind of useless .. This technique relies on the historical relative past performance of stocks and it might not perform in future.. An example of it is VOD and BT.. I have enclosed the chart of this two stocks.. Even with the benefit of hindsight one could not hedge these two stocks effectivley against each other.. Highly correlated instruments tend to shadow each other even after the spread has been created for some adverse news ..

2) Pick wo negatively correlated instruments.. The idea here is that assests rotate in the market and if for example Tech stock gets over valued the $$$ will be coming out of Tech and being injected into more defensive sectors such s Bio stocks..

3) Pick two instruments irrespective of their correlation as long as one is over valued and the other is under valued… You can use Fundamental or technical analysis to gauge the valuation // VWAP analysis which I have outlined for pair trading is the type 3


4) Pick two instruments one with high beta against a low beta instrument.. The idea is that instrument can not out perform the index all the time and need to go through the correction cycle …. Please notice that correlation does not come into the equation and it is the market sentiment which determines the correct pair.
( applicable to long to medium term only )

So why are we interested in all this techniques.. Answer is simple.. if you want to treat this game as business you must behave as a business man .. You must look into risk adjusted returns strategies than RISKY TRADES.. RISKY TRADES MIGHT PAY OFF HANDSOMLEY but there again if you give monkey a $1000 000 quid on the roulette he might bet on red and double his money too..

 

[anley]

Why are the Thought Please in operation with my 2 last posts.

They were hardly revolutionary, so why deleted?

No point sticking around boards where this kind of behaviour is prevalent.

So long fellas, I'm gone to less heavy handed boards.

 

[Grey1]

Anley,

The complete thread was deleted by error.. I am sure moderators are looking into this. ]

regards

 

[Mr. Charts]

Grey1,
What sort of annual return would an experienced pairs trader expect to make - without leverage?
I am not asking you what a beginner would make, nor you yourself, of course.
I also mean over a period of one year's trading, not an annualised figure.
BTW, I realise the relationship to risk is off the scale and every trader will be different, but a ball park figure would be appreciated if you are willing and kind enough to answer these questions.
Richard

 

[Grey1]

MR chart,

It all depends on Time frame ( intraday, weeks , month , years ) as well as the type/ style of pair trading one's undetakes.. The bench mark for hedge funds is beating a leading index but in a risk controlled manner.

As far as day trading is concerned .. It again depend on how good of a pair trader and in general a trader he is.. I would not spend 12 hours in front of terminal on a minimum wage ..

It would be meaningless to give a set figure

regards