CFD Pairs trading

[lewstherin]

Hi peeps

I'm new to the CFD world, having previously only traded/invested long in equities.
Have been doing some research into common CFD techniques and find the concept of leveraged pairs trading to be a very interesting, if marginal, means of profiting in a market-neutral sense.

My basic idea is to take a leveraged short position on something like RIO and an equal leveraged long position on a correlated stock (eg. BHP) at the end of each day. Overnight, I watch the overseas markets for an indication of the next day's move - given how the ASX seems to be the Dow's beyatch at the moment...

I adjust my stops accordingly to get out of whichever is going to take a loss, and loosen my profit taking stop on the other position to attempt to make out with a marginal profit. The idea hinges on getting out of the loss position asap, and letting the profit run just long enough to profit overall. It requires some timing to be sure, and also relies on some movement in the same profit direction after opening. I would of course be running very tight oco stops.

Most pairs trading advice I've seen seems to hinge on picking currently divergent shares that should converge back to historal correlations.
My strategy is very much a 1 day, get-in, get-out kind of trade and may work in the current volatile environment where commodities shares appear to be moving one way or another with 1-2% daily changes at present.

I'm sure plenty of folks do this, and would like to hear some comments from the more experienced folks on this?

 

[mlennox]

Never tried it, heard about it.

Have you had much success doing it on paper?

 

[michael_selway]

Hi peeps

I'm new to the CFD world, having previously only traded/invested long in equities.
Have been doing some research into common CFD techniques and find the concept of leveraged pairs trading to be a very interesting, if marginal, means of profiting in a market-neutral sense.

My basic idea is to take a leveraged short position on something like RIO and an equal leveraged long position on a correlated stock (eg. BHP) at the end of each day. Overnight, I watch the overseas markets for an indication of the next day's move - given how the ASX seems to be the Dow's beyatch at the moment...

I adjust my stops accordingly to get out of whichever is going to take a loss, and loosen my profit taking stop on the other position to attempt to make out with a marginal profit. The idea hinges on getting out of the loss position asap, and letting the profit run just long enough to profit overall. It requires some timing to be sure, and also relies on some movement in the same profit direction after opening. I would of course be running very tight oco stops.

Most pairs trading advice I've seen seems to hinge on picking currently divergent shares that should converge back to historal correlations.
My strategy is very much a 1 day, get-in, get-out kind of trade and may work in the current volatile environment where commodities shares appear to be moving one way or another with 1-2% daily changes at present.

I'm sure plenty of folks do this, and would like to hear some comments from the more experienced folks on this?

Hi is this the same idea as above?

Trade a relationship in rocky times

Back
Date: 22/6/2006
Author: John Wasiliev
Source: The Australian Financial Review --- Page: 27

Pairs trading is a useful technique for investors in contracts for difference (CFDs) when the sharemarket is volatile. Man Financial's Richard Avery-Wright says it works because it is based on the relationship between two stocks rather than market trends. It can be applied where there is an historical pattern: one share outstrips another, then the other share catches up. The idea is to take a long position in the stock that catches up and short sell the other. In May 2006, Avery-Wright says Zinifex and BHP Billiton provided a "classic" example of this type of behaviour

thx

MS

 

[Bobby]

lewstherin,

I strongly suggest you only try this with a CFD provider that uses a DMA platform, Not MM .

Bob.

 

[lewstherin]

lewstherin,

I strongly suggest you only try this with a CFD provider that uses a DMA platform, Not MM .

Bob.

Yeah, definitely.
I think the required conditions would have to include:
- DMA
- volatility
- strong leader-follower behaivior between international markets
- personal time available to monitor progress and time exits according to the day's market
- a pair of stocks that have very high liquidity, offer decent leverage margins, and track closely together.

Michael, the example you quoted above is the classical pairs trading stuff I've seen mentioned around - trading correlated shares that are divergent at present with the hope that they will converge in the near future.
My particular method relies on the pair being currently convergent and remaining so. In fact a divergence from correlated behaivior would be a no profit scenario for my method, but a profit case for your quotation.

I think I'll try some small trades to figure this strategy out...need to study up on which shares and what kinds of entrance/exit stops to set.

 

[flyhigher]

pair trading is a old tricks of big investment banks. it is out of favor in institutional market for quite a while because all arbitrage strategies opporunites are diminishing with more and particpants. Normally traders will run regression analysis for comparartive stocks. they enter the market When the prices diverge to 3 standard deviations. better run the back test before you do so.