Market Neutral Trading

Rog1111 - thanks for your reply. I'll print and read it offline where I can give it the attention it deserves. Looks interesting.

LevII - I only pair trade a trendless market. Maybe that's the key point as the two approaches are diametrically opposed (just as is an oscillating market is to a trending one!).
 
rog1111 - I'm a little confused as to why you'd want to pair trade MSFT/SBUX. My understanding of a market neutral pair is that you choose 2 stocks from the same sector so that the same news/geopolitical events affect both stocks in a similar way. e.g. 2 retailers or 2 financial. By picking two stocks in different groups or sectors you could be doubly wrong, with the short stock rising and the long stock falling, although I suspect that youd still be adequately hedged the next time someone flys a plane into a building.

As a first step, I'm going to look at correlated stocks within the same sector using "match-maker" within AIQ to find stocks with a high degree of correlation over 12 months, and then look at the ones with greatest and least RS to the sector. This should be a simple approach that should be relatively easy to back-test as well, which is a bonus.

An alternative approach is the sector rotation approach mentioned by TheBramble ( and hopefully this is what he means) where you look for strong and weak sectors, and then short the weak stock in the weakening sector, and go long the stronger stock in a strengthening sector, BUT ensuring a high degree of correlation if possible. My initial thoughts are that you should be looking at sector charts first, and looking for buy/sell signals in the sector charts before looking for pairs within the sectors.

Thanks again for bringing this up Niall - opens up a new field to explore.
 
Position size is also an important factor to consider when pair trading. For example a $20 stock is not likely to have the same ATR as a $60 stock. I would consider trading by the ratio of one to the other so that you improve your chances of a truly market neutral strategy.



Paul
 
Hi Paul

This is about position sizing really. I don't think that the stock's price is necessarily indicative of it's likely ATR when the ATR is expressed as a % of the price eg $20 stock might have an ATR of $1 = 5%, 60$ stock has an ATR of $6 (that would be nice) = 10%. Which is the more volatile ? In this case, from a position sizing point of view, the $60 stock is twice as volatile. Personally, based on previous observations, if I was trading one against the other I wouldn't want to use a $ ratio of 3:1 based on price, rather 2:1 based on volatility. But others may be happier going with 3:1 - just my opinion

rog1111

Trader333 said:
Position size is also an important factor to consider when pair trading. For example a $20 stock is not likely to have the same ATR as a $60 stock. I would consider trading by the ratio of one to the other so that you improve your chances of a truly market neutral strategy.



Paul
 
Roger

Granted, what I described is this example does not involve stocks in the same sector, rather just another Long-Short method based on :
a) the selection procedure that I outlined in an earlier post
b) volatility sizing
c) the idea that most Nasdaq 100 stocks will behave in a somewhat similar fashion in times of crisis (or elation)

Yes, I could be doubly wrong, maybe on more occasions than with same sector stocks, maybe not. The success of this approach simply relies on the effectiveness of the excel collection method at identifying a strong stock and a weak stock, based on total bid/ask sizes, and good position sizing based on volatility (ATR). Just a different approach.

rog1111

RogerM said:
rog1111 - I'm a little confused as to why you'd want to pair trade MSFT/SBUX. My understanding of a market neutral pair is that you choose 2 stocks from the same sector so that the same news/geopolitical events affect both stocks in a similar way. e.g. 2 retailers or 2 financial. By picking two stocks in different groups or sectors you could be doubly wrong, with the short stock rising and the long stock falling, although I suspect that youd still be adequately hedged the next time someone flys a plane into a building.

As a first step, I'm going to look at correlated stocks within the same sector using "match-maker" within AIQ to find stocks with a high degree of correlation over 12 months, and then look at the ones with greatest and least RS to the sector. This should be a simple approach that should be relatively easy to back-test as well, which is a bonus.

An alternative approach is the sector rotation approach mentioned by TheBramble ( and hopefully this is what he means) where you look for strong and weak sectors, and then short the weak stock in the weakening sector, and go long the stronger stock in a strengthening sector, BUT ensuring a high degree of correlation if possible. My initial thoughts are that you should be looking at sector charts first, and looking for buy/sell signals in the sector charts before looking for pairs within the sectors.

Thanks again for bringing this up Niall - opens up a new field to explore.
 
Yes, I could be doubly wrong, maybe on more occasions than with same sector stocks, maybe not.


And of course you could be doubly wrong if you just go long 2 separate stocks. At least with this method, even with pairs from different sectors you are protected to a degree from the worst excesses of major geopolitical risk which would take all stocks down with it, regardless of sector. It's just a case of how "market neutral" you want to be I guess.
 
Its good to read of the various approaches to this strategy, its definitely clear that there are numerous ways to tackle it and great to hear peoples thoughts on the topic..

I may have said this earlier, but anyway, one thing I have noticed is that as my confidence and success in my own particular genre of market neutral trading has grown.. so has the work that I put it to it.. which is right and proper really.. read so often of the daytrader mentality of being able to roll out of bed in the morning (the usa types anyway!) stumble across to the monitors... trade the first hour ("its where I make all my profits, whats the point of trading at any other time?") make a quick few grand.. and then topple back into the pit until time to go to the beach... haha

Ah well.. I was never quite like that.. way too old.. but it did seem at first as if money could be made without too much effort.. sure learned that was not the case.. at least for me.. altho I accept that there are little genies out there who can do it as natural as boil an egg.. but I do not care if I never hear another bouncy exhuberant word from that sort.. please!

For me the work is welcome.. makes me feel as if I deserve any gains I make.. and I think trading is best suited to a compulsive type of personality too.. which I guess is me.. I seem to spend a huge amount of my time thinking or talking about trading.. I am fortunate in that my wife shares the obsession and so our leisure time often involves plotting and planning and discussing.. never really get tired of that side of it... we might take the dog for a walk along the shore... and in between throwing sticks for him.. and admiring the sound of the waves breaking on the rocks and the view across galway bay to the mountains of connemara.. we will be chatting about the latest pair performance.. or better still.. which bit of the cote d'azur we will move to when we finally get this working right.. nuts right? lol
 
RogerM said:
An alternative approach is the sector rotation approach mentioned by TheBramble ( and hopefully this is what he means) where you look for strong and weak sectors, and then short the weak stock in the weakening sector, and go long the stronger stock in a strengthening sector,

Rog, wasn't my post on sector rotation. And my approach to pairs is quite the opposite to the one you state.

I short the strong and go long the weak.
 
I agree. The main point that I had been really trying to get across was that sizing on each side is a critical factor, and for my strategy at least, a volatility neutral long-short approach seems to outperform $ neutral pairs, historically.

Interesting thread.

rog1111

RogerM said:



And of course you could be doubly wrong if you just go long 2 separate stocks. At least with this method, even with pairs from different sectors you are protected to a degree from the worst excesses of major geopolitical risk which would take all stocks down with it, regardless of sector. It's just a case of how "market neutral" you want to be I guess.
 
Rog, wasn't my post on sector rotation. And my approach to pairs is quite the opposite to the one you state.


Tony - lol - That's what makes a market! :D

And in fact we could both be right provided we are using methodologies that work for us. Or of course we could both be wrong.
 
RogerM said:


Tony - lol - That's what makes a market! :D


No Rog, I understand the merriment in what you thought I was saying, but my point was I was being mis-quoted on what my pair trading approach was.

I agree, there are enough people going in the opposite direction to me to make a profit. And one day, it'll be me making the profit.
 
RogerM said:
Not wishing to go "off topic" in this thread, I've posted in the options forum a strategy which I think should be more or less market neutral.

"I'd envisage this as a bit like a fishermans night-line - set it up and leave it alone until morning - or in this case, expiry - and come back and see what you've got."

"Anyway - those are the bones of the idea - a sort of market neutral strategy where big moves are neutral and sideways markets enable you to collect premium."

Interesting post there Roger.. I have to say I understand nothing at all about options so some of it is beyond me but I get the general drift.. which seems good sense..

I quoted a couple of your comments above which connect with the way I am thinking... so far for me my methods have definitely led to a situation as you describe, altho still trying to understand the movements, generally I appear to survive big moves but make most progress when the market drifts.. which I think then allows my stock pairs to find their own direction better.. which if I have picked well, re strong and weak selections, should be in my favour..

I like the fisherman line image too.. it does seem a bit like that to me too sometimes.. once my pairs are set I tend to sit back and wait and see.. and the watery theme continues then for me too.. the behaviour of the portfolio as a whole seems to ebb and flow like a tide.. or the swell of the ocean... pairs rise and fall.. edging closer to the stop.. and then bobbing back towards the target.. each time oscillating nearer and nearer to one or the other.. until with luck I can reach over the edge and net a live one.. or cut the line and let it fall if the other applies.. lol

I am getting carried away here so had better cease.. but thanks for the interest and input... interesting..
 
Niall - your "waxing lyrical" about wandering along the beach sounds just like my own aspirations. Happy to work at the trading, but definitely need a method that allows me to get away from the screen and do other things.
 
long and short

Interesting thread. I can certainly empathise with Niall's preference for a long/short strategy.. Much less swings and the advantage of being relatively market neutral.

I have spent several months working on a long/short approach. So far I have not traded it for real but simulations show expected returns that are positive! Roughly 7-10% unleveraged.

The clear advantage of this approach is the stability irrespective of what the underlying market does. That said, the returns appear to be stable but not mind-blowing. It ought to be able to withstand some leverage though...
 
Last edited:
Top