You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Pair Trading Strategy Journal
- Thread starter jonnysharp
- Start date
Quote from Magnas:
Hello,
two traders on Friday:
Opened new pair:
Bought KIM @ 15.65
Sold WRI @ 23.20
Closed
ACE-CB for gain 1,95%
Some stats - first trade opened 5.March 2010
So far 28 trades opened, 19 closed. Win ratio 78.95%, RRR 1:1.4
Average profit 434.20 USD per trade.
Realized profit so far: 4985 USD - cca 5% on my paper capital. Not bad after about 1 month of learning. Commission would be about 140 USD.
All trades are real-time put into my broker paper account.
I am including all my trades from my journal (Exit/Entries etc).
Comments on selected pairs etc. are welcomed.
M.
Hi Magnas
Thanks for the continuous posting of your trades in this forum and the recent trades/progress summary.
You have been doing quite well with your paper account.
I see from your earlier posts that you initially began following pairs in excel.
Are you still using excel?
It seems that there is great diversity in the pairs that you watch with respect to correlation trends ...according to my metrics (I have been using PTFinder for the last 2 mos.)
Could you elaborate on what you look for in terms of correlation minimums, correlation trends, st.dev mins, % from mean, and ratio patterns?
Also what st.dev & correlation lengths you use?
...& any other tools that play in your decision making.
I have followed and researched Jonnysharp's trades & performance as well as certain others in this forum & elsewhere
.... & am still developing my own strategy.
Any information is appreciated!
Quote from Darshan:
Mergers & Aquisitions & Takeovers oh my...
Do you guys have any strategies for this? In the past couple of months I've gotten clipped by two takeover and sustained huge losses as a result.
The latest one was AREX/ME
I was long AREX short ME
This was the homework I did...
Checked all news of possible rumors of takeover prior to...
AREX is an $9.00 stock, ME is a $20.00 stock....check
AREX had a much lower price to book.....check
AREX has about half the maket cap them me.....check
and yet ME was aquired by apache.
While this is a sound strategy a lot of the anomoly moves comes when there is some news hitting the market (insider trading, news leak) and we unfortunately fade it.
I have been trading 10% a leg which in my opinion is still too much but I'm only doing it because it's a small pairs test account. Sure I'm not going to get amazing 75 % returns but I'm up 37 % in 11 months and for me thats pretty good...
Now If I was trading a larger account I would probably drop it down to 5 to 7 percent a leg. I assume that any stock I enter can go to zero ( unlikely) but the point is I only position size to where I'm comfortable losing 5 to 7 % on one trade should it go to zero.
In other words, make small trades that are in effect insignificant by themselves but have many of them and over time, as you apply your edge you will make money. This is the only way to protect yourself against unforseen events and stay in the game...
Nick
Quote from Darshan:
Mergers & Aquisitions & Takeovers oh my...
Do you guys have any strategies for this? In the past couple of months I've gotten clipped by two takeover and sustained huge losses as a result.
The latest one was AREX/ME
I was long AREX short ME
This was the homework I did...
Checked all news of possible rumors of takeover prior to...
AREX is an $9.00 stock, ME is a $20.00 stock....check
AREX had a much lower price to book.....check
AREX has about half the maket cap them me.....check
and yet ME was aquired by apache.
While this is a sound strategy a lot of the anomoly moves comes when there is some news hitting the market (insider trading, news leak) and we unfortunately fade it.
Darshan,
You may want to consider the size of both your stocks relative to the sector/industry. I consider market cap to be as important, if not more important than all the other P's....PE PB PC/F etc.
When looking at ME with a mkt cap of 2 Billion it is still considered a small cap.
We cannot eliminate the merger risk if we are holding overnight short positions...the only thing we can do is try to stack the odds more and more in our favour, so that when news like this does occur we are able to dodge these landmines. I will on occasion short a small or midcap although with less size than normal.
Mark
Hello irucken,
I am still in a process of learning and investigation so please take my experience as something that can change in the future.
I am using very basic setting for mean and std dev â I use 14 for both of them. I am not sure about the significance of actual correlation. I have done lots of tests and studied if the results distinguish when using correlation as condition or not. And to be honest, I didnât find it too useful for myself. But of course there must be some connection between the 2 stock I use for the pair. So what I do I base my pairs on their historical performance. I test the pair (using stocks from the same sector) using long history (7 years and more) and study the results from the money-management point. I use very simple entry pattern â just the cross of +/-2 std. dev and exit at mean. At the moment I am study if stop-loss or profit-target can help improve the results.
So basically I focus more on studying the trades from statistical and portfolio part than trying to âtuneâ each trade. Right now I for example try to involve money-management based on individuals betas of the pair so I take into account different characteristic of the pairs etc.
With this said. I think I am heading more the way that I will build a portfolio of pairs and then I will take every signal I get. I have a friend that has build Excel âautotraderâ and trades similar way and it looks very promising. Not only that it produces nice returns, but one really doesnât spent any time with executions. But I am not myself that far â I do the execution by hand now (and to be honest this is the weakest part of my testing as I dont have time to watch the trade every day).
Regarding tools. Yes I have my own Excel tester. It basically takes two stocks, downloads history prices from Yahoo and backtest the pair. I have build also the function that test different parameters for mean, std dev, entry and exit so I can see how robust is the pair â I like the one that are as robust as possible (that means it really doesnât matter if I enter at +2 dev or 2.5 dev). I backtest this way every pair I found interesting â for example from this forum etc. What I look for is nice equity with minimum drawdowns. I was trying to develop also âscreenerâ but for now I use the one that is available at pairslog.com as it is free and helps me find what I need.
M.
I am still in a process of learning and investigation so please take my experience as something that can change in the future.
I am using very basic setting for mean and std dev â I use 14 for both of them. I am not sure about the significance of actual correlation. I have done lots of tests and studied if the results distinguish when using correlation as condition or not. And to be honest, I didnât find it too useful for myself. But of course there must be some connection between the 2 stock I use for the pair. So what I do I base my pairs on their historical performance. I test the pair (using stocks from the same sector) using long history (7 years and more) and study the results from the money-management point. I use very simple entry pattern â just the cross of +/-2 std. dev and exit at mean. At the moment I am study if stop-loss or profit-target can help improve the results.
So basically I focus more on studying the trades from statistical and portfolio part than trying to âtuneâ each trade. Right now I for example try to involve money-management based on individuals betas of the pair so I take into account different characteristic of the pairs etc.
With this said. I think I am heading more the way that I will build a portfolio of pairs and then I will take every signal I get. I have a friend that has build Excel âautotraderâ and trades similar way and it looks very promising. Not only that it produces nice returns, but one really doesnât spent any time with executions. But I am not myself that far â I do the execution by hand now (and to be honest this is the weakest part of my testing as I dont have time to watch the trade every day).
Regarding tools. Yes I have my own Excel tester. It basically takes two stocks, downloads history prices from Yahoo and backtest the pair. I have build also the function that test different parameters for mean, std dev, entry and exit so I can see how robust is the pair â I like the one that are as robust as possible (that means it really doesnât matter if I enter at +2 dev or 2.5 dev). I backtest this way every pair I found interesting â for example from this forum etc. What I look for is nice equity with minimum drawdowns. I was trying to develop also âscreenerâ but for now I use the one that is available at pairslog.com as it is free and helps me find what I need.
M.
Quote from irucken:
Hi Magnas
Thanks for the continuous posting of your trades in this forum and the recent trades/progress summary.
You have been doing quite well with your paper account.
I see from your earlier posts that you initially began following pairs in excel.
Are you still using excel?
It seems that there is great diversity in the pairs that you watch with respect to correlation trends ...according to my metrics (I have been using PTFinder for the last 2 mos.)
Could you elaborate on what you look for in terms of correlation minimums, correlation trends, st.dev mins, % from mean, and ratio patterns?
Also what st.dev & correlation lengths you use?
...& any other tools that play in your decision making.
I have followed and researched Jonnysharp's trades & performance as well as certain others in this forum & elsewhere
.... & am still developing my own strategy.
Any information is appreciated!