Pairs Trading with Options

[wayneL]

Anyone game to try and develop this idea?

 

[Grinder]

Hmmm. Sounds interesting, Im game!

I presume one would want to be selling options instead of buying em, but being naked leaves too much risk for my liking. I'd go with spreads, my fav strategy the IC would suffice, possibly using opposing Indexes one overvalued while the other undervalued if you can find em. Otherwise, a lopsided IC with the overvalued one having a more bearish bias & the undervalued one more bullish.

 

[mazzatelli]

Assuming our cointegration models are effective I'd imagine using options for built in stop losses, otherwise execution and -edge can be cumbersome compared to spot.

I'm thinking out loud, could be jumbled

Long calls and puts:
If the pairs stay stagnant, equities can be closed out for close to nil less commisions.
Long calls and puts will have lost time value.
Probably use longer term ITM long calls and puts for deltas close to 1 and minimal time decay?

Backspreads
Risk I would imagine is in a stagnant market. The effect would be more pronounced than long vanillas. Maybe use this on very volatile tickers?
But if the pair that is meant to be "long" explodes the other way there is still a chance to profit, so both sides could profit

Verticals:
ATM debit spreads?
Credit spreads, usually one of the pairs will be in the red, so R:R of the credit spread would have to be reaaalllly good.
In built stop losses, but time decay is still working against this

I'm not sure about being short gamma for pairs trading. R:R would be cr@p

 

[wayneL]

My view is that you definitely NEED positive gamma for a pairs strategy to work.

But then there is sister theta... and then vega can open another can of worms.

The more I think about it, the more I think the idea sucks.

There's a discussion about it over at ordinarytrader. The "gurus" there haven't come up with much either.

 

[mazzatelli]

I forgot to consider vol!!

What about ATM verticals and then adjust to flies if the market decides to stagnate?

 

[mazzatelli]

Considering vol
Long Straddle for the pair to be sold. Bear deltas precipitate vol line increase.
Offset Time Decay via flies/condors [bull deltas and decrease in vol line] or calendars if IV is low

Execution and commissions would be the downside
Critique?

 

[wayneL]

Considering vol
Long Straddle for the pair to be sold. Bear deltas precipitate vol line increase.
Offset Time Decay via flies/condors [bull deltas and decrease in vol line] or calendars if IV is low

Execution and commissions would be the downside
Critique?

Disclaimer: Never done pairs trading and don't understand it well. But is seems to be overly complicating a simple concept, viz, the relationship between two supposedly correlated instruments.

Too many factors make for a chaotic system... or maybe this is just doing my head in.

 

[mazzatelli]

Disclaimer: Never done pairs trading and don't understand it well. But is seems to be overly complicating a simple concept, viz, the relationship between two supposedly correlated instruments.

Too many factors make for a chaotic system... or maybe this is just doing my head in.

The pairs need to be cointegrated, but not necessarily correlated.

Too much commissions, contest and execution risk with options. I could not imagine trying to carry this out on ASX ETO's

 

[Grinder]

Disclaimer: Never done pairs trading and don't understand it well. But is seems to be overly complicating a simple concept, viz, the relationship between two supposedly correlated instruments.

Too many factors make for a chaotic system... or maybe this is just doing my head in.

Me too. It's not making much sense, hunted around for some info on the subject but did'nt come up with much of anything that would make me want to put anything on. Options have enough variables on their own to keep me on my toes, don't need to complicate it anymore than what it is.

 

[mazzatelli]

I was brainstorming, but maybe clarify what I put up in case others checked out the thread.
Btw not a big fan of stat arb

Long calls and puts - Long call as a substitute for long stock
- Long put as substitute for short stock

Backspread - Call backspread as substitute for long stock
- Put backspread as substitute for short stock

Debit spread - ATM bull call spread for long stock
- ATM bear put spread for short stock

NB: Just ideas, not proven methodology

 

[mak jeng]

Pairs trading with options probably requires some very special situations.
I think it would probably introduce hard to manage risk in the trade.
I do a few pairs trades via CFDs e.g. RIO with BHP.
As suggested, a long straddle but with a fairly long duration might pick up anticipated (differential) price movement in excess of the cost of time decay. The holding cost with CFD is very much less and execution is far simpler.

 

[sails]

Bit late to the party on this one - but wouldn't a synthetic short (sell call/buy put) on one and a synthetic long (buy call/sell put) on the other do the trick? Or have too many cobwebs taken over my brain?

 

[lukeaye]

have contimplated this one,

This is an old thread but here goes anyway.

From what i have discovered, the advantage of pairs trading is the fact you dont have to know what price the stock is going to go to or even the direction of the trend. Your probabilty of success is high, but the trade of is usually a limited return.

Using options you haev break even pionts, to pairs trade effectively you have new added a new dynamic, now both stocks have to move past your breakevens. So this would limit your profitibailty drastically, to the piont where it just wouldnt be worth it, especially with the price of aussie options.

I dont see how this can develop into a particurally good trading system. for me, the edge is taken away

 

[mazzatelli]

ugh...6 months on...
Idea was sprung of a fellow colleague recently who trades these on bear index signals and corr. spreads - still subject to personal back-testing

Ratioed pair in favour of component with lower iv, translating to upside gamma curvature + long vega/gamma - to compensate for corr./coint risk

OTM puts so that prints into the index skew contaminate vol levels [short component] and long component IF corr. holds will benefit due to > ratio.