Pairs Trading

MR CHART ,

I am not attacking you .. I Simply donot think you should be preaching what you dont understand .. As I said before when every body was attacing you I was quite..

My interest is in your technical views which you choose NOT to reply .. Donot you think you owe that to the members of this BB
to simply explain why u think ADBE had a LOW RS when you simply donot use historical evidence ? ( ADBE is up by 46C while DOW only risen 13 points )


If you choose not to reply to my technical question then I wont bother you .. I am sorry being aggressive towards you , this is the habit I have picked up being an X- research fellow in the finance industry ..

Regards
 
Last edited:
Pair Trading is new to me but I am becoming interested in this.

How would one go about identifying suitable stocks for pair trading and where can I go to get an insight into the know-how required for this type of trading and the mechanics of it ?
 
Swing Trading Pairs

I have recently begun a market neutral pair trading strategy.. swing trading 5 pairs at a time at the moment.. and was glad to find some other's experiences here about the general method... I have daytraded US stocks for the last couple of years but have found a certain amount of unexpected relaxation doing this compared to my daytrading years.. not that anything about trading can ever really be called relaxing! lol But after the tenseness that was hard to avoid in intraday trading I really do like this.. and altho my goals are more modest these days I think there is a lot to be gained from such a strategy.. so long as, as someone mentioned here earlier, you are doing it right!

Thus far, I am using basic TA pairing.. long strong and short weak stocks within a sector.. intending to increase to ten pairs over time.. I shall investigate further the more complex corellation methods I have heard described but so far this is seeming to be working quite nicely.. so long as the stocks have nice patterns and each relate to their sector and each other in a pleasing way..

To do this in as low risk manner as reasonable is clearly not going to produce the kind of results that intrepid daytraders have learned to aspire (perspire) towards... but I have been pleasantly surprised by the stable nature of the actions of the pairs so far.. its new to me to regard two stocks as one.. and be looking not at the individual performance but at the combined P/L of each.. I am not using the ratio charts much but prefer to see each stock in its own right for the ta aspect and a normal comparison chart.. sierracharts spreadsheets allow really nice live analysis of the performance too.

Neil (Niall)
 
Niall,
A very nice post and very similar in many respects to what I've been doing very successfully for a while now.
Well done - perhaps you might agree it's relatively simple to choose stocks and to trade this way.
In trading it's what works that matters.
Good trading
Richard
 
Niall,

Glad to see another pair trader ..

Pair trading is the most relaxed way of making $$$ but needs as you said a knowedge of sectors and correlation analysis of stocks based on their historical performance other wise it can not be called a pair trade..

Study of Correlation analysis and its application intra-day is some thing that I am not quite sure is the way ahead . This is because we expect past to strongly relate to the future on the day we take the trade and this does not always happen ..However in longer time frame the picks are sort of easier..

There are however other techniques to pair trades stocks intra day which I am going to discuss in a private forum based on VWAP analysis which are easily implementable and has proved to give more risk adjusted return than any other methdology I know..

I have used Correlation analysis to pick divergent /convergent stocks for many many years for swing trading but found I had a poorish overall result hence moved to some thing more efficient..

MR CHART ,

You have never chosen stocks based on sector correlation and never used Correlation analysis . Niall, method is nothing like yours.

All you do is choosing two stocks one going down the other going up 5 minutes into Open and call it a pair Trade :LOL:
 
Generally, what percentage success rate does pair trading give ?

ie. percentage of total trades that are successful taking a pair as one trade.
 
salty,

I am sorry but I can't put it into %.. What matters is that you have a market neutal postion where spike's or break outs wont affect your overall postion to an un-tolarable limits..

I have been long VRSN and short COST well before green span and even the sudden drop in DOW did not affect my overall risk ..


regrds
 
Thanks for the replies.. I am not sure myself of which methods of pairs trading are better than other.. just carefully working myself into the strategy so far.. but as I said above I like it a lot so far..

Also it seems to be there must be many similarities but also many differences between intraday pairing and the swing pairs I am trying.. and if intraday pairing I presume you are mostly sticking with one pair at a time? I have tried managing multiple positions intraday before and while it is certainly possible it must take a lot of skill and practice.. depends on if veering towards scalping or more intra day swings I guess..

A key element of the portfolio based swing pairs to me is the smoothing out of the action that having the capital spread across a number of pairs can hopefully create.. especially come the day that a major event hits one particular stock.. the best laid pairs can fall flat on their face in that case.. the portfolio, I would think, is better placed to weather such a storm..

Niall
 
Another of my pair trades:
entries:
IACI long 33.05 15.07
(relative strength and about to break through y'day's high plus cleared whole number)
ICOS short 38.85 15.02
(relative weakness)
exits:
IACI sold 33.62 16.33 for profit +57
(move appeared to falter)
ICOS covered 38.69 16.36
(move faltering) for profit +16
total profit +73

The ideas are exactly the same and very basic - going long a strong stock and short a weak stock. Additional factors such as those I mentioned on the IACI entry add weight, of course. It does help to have the experience of reading time and sales and and also level 2 to a lesser extent for entries and exits.
You can trade like this in the opening minutes but you do require a lot more experience and I don't suggest anyone tries it to begin with.
The trades after 3pm UK are easier as direction tends to be clearer.
IMO people should paper trade this type of play many times first to get the hang of it before risking a single dollar.
HTH
 
Here are the charts of the pair trade between SYMC which I went Short at just after 3.00pm and YHOO where I went Long less than 30 seconds later.

I have to admit that I didnt manage this trade as well as I should have done because I was getting so many signals that I didnt confirm some of the things that I usually do particularly on the Level II screen. Anyway the trade made a profit of 43 points in total with the SYMC trade making +47 and YHOO losing at -4

If you pair trade always put your short trade on first because some days it is not possible to short a stock and if you have already gone Long and cannot get your Short position on then your risk is greatly increased.


Paul
 

Attachments

  • symc 16th jan.png
    symc 16th jan.png
    7.7 KB · Views: 612
Hi Trader333,

A couple of questions if I may, regarding the YHOO long:

- What were you looking for on Level II?

- What kind of stop were you using? Just curious as you let the trade go against you quite a way and didn't stop out.

Thanks,

Proxy
 
proxy,

The idea of a pair trade, (when I do one at least) is that it is market neutral so although one stock may move quite a way against you the other one counters this and your overall position should at worst be flat or within stop limits for the joint position.

Once I am in the trade I look at my overall profit and loss and not at how each individual trade is performing. If my overall position had gone against me by around 25c for this trade I would have closed both trades simulataneously but that didnt happen and the potential reward was over 50c so a 2 to 1 Reward to Risk ratio. On the Level II screen I look for Key MM support on the bid for entry on the Long trade and resistance on the Ask for the Short trade again from key MM. I tend to use Level II more for optimising entry when pair trading more than anything else nowdays. When I am taking directional trades then I am glued to the level II screen throughout the whole trade.

On the pair trade I just didnt get the best entry that I could have done if I had used Level II properly which cost me around 10c in profit across both stocks.


Paul
 
Paul
Do you program the ratios between stocks before the trade & in rt or is this a manual process please
Ron
 
Hi Paul,

Thanks for your reply - that makes a lot of sense.

I see you're using VWAP +/- 2 std devs as a means of picking your longs and shorts. Do you use anything else when picking your stocks or is it just that at that particular instance YHOO and SYMC were at opposite extremes from the VWAP (i.e. both stocks are likely to reverse)?

Regards,

P.
 
Paul
What I meant to say is that when you have identified potential pairs to trade do you mechanically match the r/r ratios. While managing the trade do you use say excel to show your net position or do you just calc it manually.
It does interest me to know how you decide on the amount of stock to trade on each side as you must strike the best balance to achieve a market neutral result - am I making any sense?
 
proxy,

I scan over 160 stocks continuosly througout the day. At the time it is just as you have said, although I had too many signals all at once which did distract me but I picked 2 that I thought would give the best chance of reduced risk.


Paul
 
Ron,

My net overall position is displayed in IB during the trade so that is not a problem. I am not sure how others have set things up but I have programmed in a number of factors into my alerts which mean that the profit potential in any trade I am alerted to is not less than 35c for each stock. On days like today where I was getting a lot of alerts I have set it even higher as it is a parameter that I can easily change. So based on that I dont match each reward to risk ratio but I do make sure that it is not less than 2 to 1 which takes less than a couple of seconds to determine. As with anything in trading I will be refinig this as things progress and may be get even better at reducing risk which is of prime improtance to me in any trade that I take.


Paul
 
You must log in or register to reply here.

Similar threads

Pat494
Terrorism
2 3 4
Replies
69
Views
13K
tomorton
T
S
Gordon's visit to the troops.
2
Replies
38
Views
5K
sinimini
sinimini
C
pairs strategy
Replies
14
Views
3K
jamezz34
J
O
Estimating Option Prices
Replies
15
Views
3K
Martinghoul
Martinghoul
L
Cheaper to trade,YM or DIA?
Replies
3
Views
3K
roguetrader
R
Top