Pair Trading

Riskyman

Well-Known Member
#11
About the opportunity cost of holding a pair, we break down our trading system to trade 4 types of signals.
1. Very Short Term ( 1-2 Days or even Intra Day)
2. Short Term ( 3-7 Days )
3. Positional ( 5-20 Days )
4. Tracking the Structural Break

The days mentioned are working days, obviously. Now the time frame which is the highest is obviously the 4th one when a pair undergoes structural break. We obviously dont know the reason behind the structural break until very late ( when we read about it in media ). So the opportunity cost is highest in the 4th category.
Thats the problem!. A lot of times price does not act like we want it to. A few days ago, I took a very short term trade and my calculations said the pair on a certain time frame should revert to mean in less than 4 trading days. Ended up coming together on the 8th day. The 4th month long wait i mentioned was meant to be a positional trade that ideally should have fructified in about 30 trading days.

Take a look at Ranbaxy and Sun pharma. They dont really offer much of a trade on daily TF. But on small time frames like 10-30 min charts you can find prices tend to sway. The beauty of this pair is when you find a trade, you can always make money.

Another question out of curiosity... Do you trade pair with stop losses?
When im pair trading, I dont usually have stops, even mental. I just make sure I have enough margin and when i see some MTM losses on the board, I just swallow hard and sit tight. Mostly, they end up being profitable eventually.
 

Riskyman

Well-Known Member
#12
Big Short,

My life is all about travelling this beautiful world, meeting people and experiencing life the way its meant to be. As I get older (not old :D) I want to do more of these fulfilling things than be seated in front of a computer for 8 hours. Pair trading is a great way to make money and it doesnt even require you to monitor the pair every few minutes. Some day I plan to put an algo together. The machine can take entries/exits and churn some cash while I go chasing pretty gals.

Until then Ill live on this forum and chew some brains:thumb:
 

ethan hunt

Well-Known Member
#13
Thanks.
I do run my own excel sheet. Correlation can be done in excel easily. Its the co-integration part which takes time. Co-integration can be calculated by 2-3 different methods, specifically ADF test ( Augmented Dicky-Fueller Test ) or by Johansons Test and using OLS ( Ordinary Least Square ) as well.

Although the theortical papers and books tell us that co-integration is a must, it actually is a hindrance while doing pairs trading. I had run Co-integration test on the entire F&O universe ( well almost the entire F&O) and came up with no pairs which were fully integrated. Hence i do not use co-integration. I rarely look at correlation as well. I ll explain it in future posts. Hope to see you around here more :) and learn from your experiences as well
Which software / utility do you use to run co-integartion for several stocks in one go ?
 
#14
Another question out of curiosity... Do you trade pair with stop losses?
When im pair trading, I dont usually have stops, even mental. I just make sure I have enough margin and when i see some MTM losses on the board, I just swallow hard and sit tight. Mostly, they end up being profitable eventually.
Ok, So the first thing anyone will tell us if we are TRADING ( not investing ) is to have our Stop Losses Defined. But when we are doing mean reversion trading we cant really have stop losses, cause we cant predict how much a pair will deviate from its mean before reverting to the mean.

In my initial years , i used to do pair trading without any stop losses. So whenever a pair used to undergo a structural break like i mentioned, it would result in huge MTM losses and half the time it would not revert back to its mean. Since then i have tweaked my system so as to trade it with Stop Losses.

Now , out of the four types of signals i mentioned in the earlier post, i use stop loss in 3 of them. In 2 of the signals the stop loss is in terms of price i.e. if the spread or ratio goes beyond a certain point we exit it. In 1 of the signal the stop loss is in terms of the holding period e.g if the pair is expected to revert/converge in 2 days and it does not , we get out.

The only trade in which i dont keep a stop loss is a positional trade. Hope that answers your query :)
 
#15
Which software / utility do you use to run co-integartion for several stocks in one go ?
Although there are softwares available which will let you run co-integration test on multiple stocks in a single go, most of them are paid.

I basically use an excel model which calculates it but you can only run it on one pair i.e 2 stocks at a time. It is available for free and if you want i can give you the link for it, if its not an infraction of the rules of this forum.

But as i said, its more of a hindrance rather than help. :)
 

mastermind007

Well-Known Member
#16
Although there are softwares available which will let you run co-integration test on multiple stocks in a single go, most of them are paid.

I basically use an excel model which calculates it but you can only run it on one pair i.e 2 stocks at a time. It is available for free and if you want i can give you the link for it, if its not an infraction of the rules of this forum.

But as i said, its more of a hindrance rather than help. :)
Hi Big Short

Yes you can post links ... Usual convention is to upload it on some file sharing site (mediafire, google docs, etc etc ) and then paste the URL here
 
#17
One of the most important thing while doing pair trading is Capital Requirement. You cant do pair trade with 1-2 Lac INR. So for those of who to whom this system appeals, please only begin trading with a large Capital. I know, i know, it goes against the convention knowledge of trying out new strategies with less money, but in this trading system you need ample liquidity. Ideally a capital of INR 10 Lacs + INR 5 Lacs for drawdowns is sufficient to start with and generate substantial returns in absolute terms.

Money management plays an important part as we will be running multiple positions and multiple trades at the same time. Lack of money management led me to liquidate my positions in loss many a times during the testing phase.
 
#18
@Big Short

First: Is there any thing you can tell the forum which we do not know already? I have gone through your thread in which you even tell people that you will answer certain questions after your meal and even five days later there is no answer to the question. Just one example. Kindly correct me if I missed it.

Second: Are you further planing to move on with your thread in this way or are you able to post real posts about pair trading like posting charts, statistics and so on to educate people here or what is your main plan to move on with your only "Already known words post" for the coming future?

Kindly wait for your most interesting and most educational post you will do next. :thumb:
 
#19
@Big Short

First: Is there any thing you can tell the forum which we do not know already? I have gone through your thread in which you even tell people that you will answer certain questions after your meal and even five days later there is no answer to the question. Just one example. Kindly correct me if I missed it.

Second: Are you further planing to move on with your thread in this way or are you able to post real posts about pair trading like posting charts, statistics and so on to educate people here or what is your main plan to move on with your only "Already known words post" for the coming future?

Kindly wait for your most interesting and most educational post you will do next. :thumb:

The purpose of this forum is for the people to get their questions answered on pair trading and as my journal for the pair trades.

Someone asked a question, i answered it. Read the posts, i explained why i follow somethings or use certain things and more importantly, why i do it in the model. Then we discussed on certain aspects like co-integration and stop losses. Do i have to summarize it more to you? Someone asked, i replied. Dont read it if you already know it.

I could have vaguely answered his question regarding "buying/selling nifty or banknifty" but i thought it better to have certain discussions, establish how i am doing it so that the flow of information is better and more efficient.


Regarding your second part, I was contemplating on how to move about it. Because anyone can look at historical data and say this should have been bought or sold. So past Trades dont matter. I might be a bit busy till march ending, however i ll try to post something which matches your expected level of information soon.