Long AEC @ 5.84
Short HME @ 35.20
Short HME @ 35.20
Pair Trading Strategy Journal
- Thread starter jonnysharp
- Start date
Article from trader magazine talking about professional pair traders making heaps from the recent volatility
http://www.tradersmagazine.com/news/-103588-1.html?ET=tradersmagazine:e268:52965a:&st=email
http://www.tradersmagazine.com/news/-103588-1.html?ET=tradersmagazine:e268:52965a:&st=email
Hello,
I was wondering from the .jpg that johnny posted what the following (4) calculations
1) Ratio for the last 150 days?
2) RSI for the last 150 days?
3) Plus/Minus for the last 365 days
4) Spread for the last 50 days.
Here is what I think they are I'm sure it's not all right...
1) Is this simply FirstStock / Second stock?
2) This is basically instead of doing an RSI on price, this is an RSI on spread?
3) Not sure what this is
4) this is First stock / Ratio - Second stock
I was wondering from the .jpg that johnny posted what the following (4) calculations
1) Ratio for the last 150 days?
2) RSI for the last 150 days?
3) Plus/Minus for the last 365 days
4) Spread for the last 50 days.
Here is what I think they are I'm sure it's not all right...
1) Is this simply FirstStock / Second stock?
2) This is basically instead of doing an RSI on price, this is an RSI on spread?
3) Not sure what this is
4) this is First stock / Ratio - Second stock
1) Left stock / right stock
2) RSI of the ratio
3) Standard deviations from the mean
4) Highest stock price - lowest stock price
2) RSI of the ratio
3) Standard deviations from the mean
4) Highest stock price - lowest stock price
Yes I read a research paper saying the same, I have yet to test this theory, I think it mostly explains why profit from pair trading exists oppose to a method. Some stocks have less liquidity than their peers, therefore taking longer for industry/market wide news to be priced in. Also some institutions may trade off news whilst others won't. The results of the next trade you take is largely random, however 50 trades isn't. Small edge x leverage = profits
In fact the main conclusions from the study are:
-> The larger the value of the industry information diffusion measure, the larger the difference of individual stock's speed of response to industry common common information within the pair, and the larger the profits from pair trading. The impact of the difference of individual stock's speed of response to industry common information is paritculary strng among less liquid stocks, stocks with fewer common institutional holdings or analyst coverage
-> The level of liquidity and short-term changes in liquidity ("liquidity shock"), contribute positively to the total profts - which arise becasue of an increase in opening probability, a decrease in horizon and divergence risk, and an increase in convergence seed. However, they are also associated with increases in arbitrage risk
My interpretation is that, stocks liquidity plays an important role in pairtrading, affecting positively profits and open probability and decreasing horizon risk and divergence risk.
A less liquid stock tends to react slower to industrial news and market fluctuations.
On other hand, less liquid stocks, tends to have higher bid/ask spreads, increasing the arbitrage risk.
I would like to know, if any pairtrader have done this correlation between pairtrading and stocks liquidity.
Best Regards
I've reconsidered. And, like I have said before, I thought I made a mistake once, but I was wrong. Once again I believe that financials are a great profit center. (Read the article Jonny linked to: http://www.tradersmagazine.com/news/-103588-1.html?ET=tradersmagazine:e268:52965a:&st=email ) Volitility is swell and financials have it.
1) Financials (banks and insurance) have the largest profit histories over the last year of any group of pairs I've backtested. Does anyone know of better profit histories? See Jonny's recent 4 day $2500 success with PUK/AXA (life insurance).
2) The downside appeared to be the greater possibility for news on only one of the stocks which would negatively affect the profitability of the trade. But over time the probability washes out, and news may even be an advantage if handled properly. First, the probability that the news will be good or bad is 50/50 (or maybe 40/60), and the probability that it will affect your long or your short side is again 50/50 for each. So, over a period of time, those probabilities appear to be a wash.
3) A review of the back testing of profitable pairs shows that 90% of the trades in the financials (over $200 profit with $2500 on each side) are very profitable.
4) If you monitor your trades, you can manually get out of the adversely affected side of your trade early, and let the other side run for profit. The backtester shows high profit success with strictly mechanical trading, but we can cut the losses by monitoring the news and closing our losing positions early when we discover adverse news.
So, the bottom line of my reasoning: I'm wanting profit. Over the last 365 days, and the nearer term, properly filtered Financial pairs have shown great profits with very good win/loss ratios. What's to argue against the history?
Like I said, I thought I was wrong once, but I now think I was mistaken.
Walt B
Both last posts make the point:
Greater profits = Greater risks
Whether it be liquidity or news risks, sometimes these are one and the same.
This is not always the case of course but a pretty general rule in markets. So how would you counter this?
In general, in an increasing volatility regime you can decrease your size. If you do this you are giving up some profits to reduce your risk.
The alternative is to increase your frequency. You can do this by adding pairs or going to a shorter time-frame.
Greater profits = Greater risks
Whether it be liquidity or news risks, sometimes these are one and the same.
This is not always the case of course but a pretty general rule in markets. So how would you counter this?
In general, in an increasing volatility regime you can decrease your size. If you do this you are giving up some profits to reduce your risk.
The alternative is to increase your frequency. You can do this by adding pairs or going to a shorter time-frame.
Hey guys,
I got me an IB account ! (no easy task being in Australia)
So now I can get those nice low commissions which you guys have been getting all these years...
I'm after some advice. Since IB offers a vast array of order types compared to what we "enjoy" here on local platforms, I would like to check with you what you consider the best order type for pairs trading.
First, a little background, up to recently I was catching the last half hour of the USA market which is early morning my time. Now day light saving has ended for us , so physically trading the last half hour is a bit of an ask now
So, I'm now looking at analyzing my pairs thru my day and placing order via IB fo rwhen the USA market opens late night my time.
Under these circumstances, would you recommend just market orders for both long and short legs ? I see IB has an order type "mkt at opening" but I would have though this is basically the same as a mkt order anyway.
Personally, I don't think limit orders would be appropriate as I don't want to get up in the morning to find only one leg of the pair has been executed.
So just after any advice you guys can give, still trying to find my way around the IB platform as it's quite complex compared to what I've been used to.
I've seen written somewhere that combo orders are good for pairs, but I'm still checking this out.
Johnny, I think you use MOC orders ?
Any reason why you don't do you trades at the opening ? (sorry if I'm wrong)
regards
Ivan
I got me an IB account ! (no easy task being in Australia)
So now I can get those nice low commissions which you guys have been getting all these years...
I'm after some advice. Since IB offers a vast array of order types compared to what we "enjoy" here on local platforms, I would like to check with you what you consider the best order type for pairs trading.
First, a little background, up to recently I was catching the last half hour of the USA market which is early morning my time. Now day light saving has ended for us , so physically trading the last half hour is a bit of an ask now
So, I'm now looking at analyzing my pairs thru my day and placing order via IB fo rwhen the USA market opens late night my time.
Under these circumstances, would you recommend just market orders for both long and short legs ? I see IB has an order type "mkt at opening" but I would have though this is basically the same as a mkt order anyway.
Personally, I don't think limit orders would be appropriate as I don't want to get up in the morning to find only one leg of the pair has been executed.
So just after any advice you guys can give, still trying to find my way around the IB platform as it's quite complex compared to what I've been used to.
I've seen written somewhere that combo orders are good for pairs, but I'm still checking this out.
Johnny, I think you use MOC orders ?
Any reason why you don't do you trades at the opening ? (sorry if I'm wrong)
regards
Ivan
It's an old adage but it applies - look to let profits run and to cut losses (don't give up profits either). Look to capture as much of the move as possible not some fixed, arbitrary profit point.
B/A spreads are a fact of life and obviously they're going to affect you. It's part of the cost of doing business. Deal with it
Make sure you're using a low fee broker. At 1/2 a ct a share, commissions are a mimimal consideration. You're main focus should be on the strategy. Trade only in size that you're comfortable with.