Trading individual legs off of a pair trade?

[fika]

You are doing better because the market has been going straight up. If we get a downdraft, you will wish you hedged your strategy. If you choose not to hedge, I would consider smaller overnight position.

I agree as the bull market makes everyone look brilliant. Time will tell. I did manage to make some good trades though in the first half of May(some that I still am holding now). Only time will tell. In my experience, my portfolio tends to see some of its best days in more bearish times, and tends to lag during more bullish markets.

 

[fika]

I am statarb guy... Actualy my system works the way, that it ususally trades pairs, but it can also trade long only and short only individual stocks - all beta hedged...

Are you manual or automated?

 

[fika]

Just my humble opinion but i think you have found a method that is working in this market. You are are punishing the shorts by selecting the most oversold of peers
(Spreads becomimg more oversold) and dropping the hedge.
Can this selection method be tested in a different market environment?
Of course say/state if i am missing the point and maybe more experienced will join in.
Good trading to you.

I almost want to see a bear market come through(as we really could use one anyways) and test my portfolio. I do this all manually, studying correlated names of a stock I am interested and setting my own ratios. I then create a massive watchlist of all my spreads. At times I even spread my stocks against indices like SPY to see how a name gauges against the overall market.

Yes I think it has to do with shorts realizing they are overpaying as the spreads they watch go out of favor for them. I also think that a lot of experienced institutional traders are more willing to trade just one leg and leave off the hedge as they are thinking the same thing I am. Although this may disappear once we see more bearish sentiment kick in.

 

[tom_czr]

Are you manual or automated?

Automated. But since I am full time trader I am around computer during the day anyway, so basicaly it can be done manualy too - I have software which automaticaly construct best possible pairs based on my needs and it graphicaly shows signals too if manual entry/exit is needed (in case there will be some trouble with broker's API or something like that)...

BTW If you construct pairs correctly, there is no need to trade long only leg, because short leg will actually lower your overall margin (if you use portfolio margin account) and short part of my system is making money too - though much smaller amount, but in case 2008/2009 it is going to make brutal profit (again )

So long story short - it is possible to have portfolio insurance which is not costing money (like buying put options), but which actually makes money...

 

[fika]

Automated. But since I am full time trader I am around computer during the day anyway, so basicaly it can be done manualy too - I have software which automaticaly construct best possible pairs based on my needs and it graphicaly shows signals too if manual entry/exit is needed (in case there will be some trouble with broker's API or something like that)...

BTW If you construct pairs correctly, there is no need to trade long only leg, because short leg will actually lower your overall margin (if you use portfolio margin account) and short part of my system is making money too - though much smaller amount, but in case 2008/2009 it is going to make brutal profit (again )

So long story short - it is possible to have portfolio insurance which is not costing money (like buying put options), but which actually makes money...

Interesting, yes I get that the short side will maintain margin. My goal with this strategy is to steer clear from anything that has to do with being market neutral even if it means losing the extra margin/insurance from the hedge. I want to speculate and accept the risk lol.

 

[tom_czr]

Interesting, yes I get that the short side will maintain margin. My goal with this strategy is to steer clear from anything that has to do with being market neutral even if it means losing the extra margin/insurance from the hedge. I want to speculate and accept the risk lol.

Why not, there is not much risk being long only if done properly...
What method do you use for construction of pairs?

 

[eurusdzn]

Do you find that a simple long and also a reverse/mirror image short system
can both be profitable in even a strongly trending market?
I think this is possible because one can construct a basket of spreads that do not coint and corr closely so at any time a long and short system can work together on the same instruments at the same time. Not equally well but market filters help weigh.
Volatility is a choice as well. Coint. pairs have volatility less than individual legs but
high volatility can be found by selecting legs that do not follow each other.
Seems that basic trend and range trading instruments can be selected at the extremes of cointegration. Range is high coint and trend is no/low coint.

The above refers to etf's and i find that for a retail amateur it is hard to put on both legs and get fills due to difficulty shorting and /or illiquid options.
I guess BTFD.

 

[NKNY]

Does anyone ever model spreads/pairs and only trade one of the legs? I have been doing it for the past 2 months and it has been working really well. I just wanted to see if others incorporate such analysis into their strategies.

The idea of the strategy does make sense as you are buying an equity that is oversold from the perspective of its relationship with similarly correlated equities. So if I find a company I like from a fundamental perspective; I will make some spreads using similar companies and only go long that stock when that spread is oversold. It has helped tremendously with my entry prices that I take in a stock.

What does ET think of such an approach? Maybe this could be a helpful tweak to your strat or is it rubbish?

Hi

This can work in an uptrending market but if we have a correction you should be hedged.

With pairs you can't assume the oversold will go up, the overbought may go down.


Nick

 

[NKNY]

Automated. But since I am full time trader I am around computer during the day anyway, so basicaly it can be done manualy too - I have software which automaticaly construct best possible pairs based on my needs and it graphicaly shows signals too if manual entry/exit is needed (in case there will be some trouble with broker's API or something like that)...

BTW If you construct pairs correctly, there is no need to trade long only leg, because short leg will actually lower your overall margin (if you use portfolio margin account) and short part of my system is making money too - though much smaller amount, but in case 2008/2009 it is going to make brutal profit (again )

So long story short - it is possible to have portfolio insurance which is not costing money (like buying put options), but which actually makes money...

Hi Tom czr

Curious what length is the mean you trade off of. Are you short term or longer like 50, 100, 200 etc...


Also is your software proprietary or off the shelf.


Thanks.


Nick