If an investor wants exposure to the gold/eurusd implied vol spread (betting it will converge) would the appropriate strategy consist of long eurusd straddles and short gold straddles done in vega neutral amounts?and if you do it via vol/var swaps how would you determine the proper notional amounts?

Yes, if convergeance were expected quickly. If covergance expected to take a long time you also need to delta hedge, esp with a trade like this one. Strangles therefore might make life easier instead of higher-gamma straddles which would have you constantly rebalncing.

thanks vesel. would doing them vega neutral be sufficient to focus the exposure on the relative moves?

If you are just vega neutral, any profits will be a function of both chnages in vol and index. You therefore need to delta hedge to remove the effect of index changes leaving you as a result exposed just to vol changes. Using straddles, because they're ATM, have higher gamma, so it will mean rebalncing more often. You could therefore could use strangles which are OTM and lower gamma = lower maintance.but as i said depends how long you intened to hold. If convergance expected within a few days, probably delta hedging may not be of much benefit at all (then again its gold you're talking about so i dunno). But if its expected to drag on then you really should.

does anyone have a list of some of the main papers in this area (pairs trading)??

QuoteOriginally posted by: playerdoes anyone have a list of some of the main papers in this area (pairs trading)??Here's a decent list of some references from another post. "A Computational Methodology for Modelling the Dynamics of Statistical Arbitrage" Andrew Neil Burgess 1999 Phd thesis"The Cointegration Alpha: Enhanced Index Tracking and Long-Short Equity Market Neutral Strategies" ISMA Discussion Papers in Finance 2002"Testing Market Efficiency using Statistical Arbitrage with Applications to Momentum and Value Strategies" S. Hogana, R. Jarrowb, M. Teoc*, M. Warachkad 2003"Statistical Arbitrage and Securities Prices" Oleg Bondarenko University of Illinois at Chicago"Inference And Arbitrage: The Impact Of Statistical Arbitrage On Stock Prices" Tobias Adrian MIT"High Frequency Pairs Trading with U.S. Treasury Securities: Risks and Rewards for Hedge Funds" Purnendu Nath London Business School 2003"Cointegration and Asset Allocation: A New Active Hedge fund Strategy" Carol Alexander 2001

anyone have a copy of "A Computational Methodology for Modelling the Dynamics of Statistical Arbitrage" Andrew Neil Burgess 1999 Phd thesis. I can't locate it online. henderson.geoffrey@gmail.com

QuoteOriginally posted by: Hendersonanyone have a copy of "A Computational Methodology for Modelling the Dynamics of Statistical Arbitrage" Andrew Neil Burgess 1999 Phd thesis. I can't locate it online. henderson.geoffrey@gmail.comSee attached.Good luck!

kitchenware,hi,do you have something from this list or can you get a link?"The Cointegration Alpha: Enhanced Index Tracking and Long-Short Equity Market Neutral Strategies" ISMA Discussion Papers in Finance 2002"Testing Market Efficiency using Statistical Arbitrage with Applications to Momentum and Value Strategies" S. Hogana, R. Jarrowb, M. Teoc*, M. Warachkad 2003"Statistical Arbitrage and Securities Prices" Oleg Bondarenko University of Illinois at Chicago"Inference And Arbitrage: The Impact Of Statistical Arbitrage On Stock Prices" Tobias Adrian MIT"High Frequency Pairs Trading with U.S. Treasury Securities: Risks and Rewards for Hedge Funds" Purnendu Nath London Business School 2003"Cointegration and Asset Allocation: A New Active Hedge fund Strategy" Carol Alexander 2001

over the years, I have seen numerous posts on stat arb. every now & then some fresh kid of school writes a new post for info on stat arb. in so far as pure equity arbitrage, in my Fund of Funds job, I have met nearly all the dedicated statistical arb funds out there. the academic papers (e.g. cointegration etc) do not work in practice; too less alpha, too volatile. there are periods when it does well and when it does poorly. in each cycle more funds go dead than new funds launched subsequently. the universe of live statistical arb funds has been shrinking - currently only 6-8 firms manage more than $100mn in pure statistical arb & perhaps less than around 20 funds out there that do dedicated statistical arb. it has become quite a small space by 2009. the alpha has disappeared. i haven't come across any recent successful ones: the returns over multiple years has been averaging sub 7-8% and going down steadily. 2008 was an occasional 8-10% year, but most of the '00 decade the typical fund returned 5-6%. to cut a long story short: majority of the funds trade some sort of stock vs a replicating basket, a small minority play basket vs a replicating basket. the surviving ones mostly play the former category. most of the funds that go regularly dead play the convergence, while some of the surviving ones play both the divergence & convergence. the vast majority of quant equity funds are the low to ultra-low frequency fundamental information based funds, e.g. many that play value: long high value/earnings, short low value/earnings etc. most of the pair trading firms, which are actually highly successful compared to stat arb funds, purely use fundamental information about the stocks and not based on statistical relationship. another tiny number of higher frequency equity arb funds are in the category of news flow driven, highly successful, but again not pure statistical arb. another category that has mostly gone dead, similar to pure statistical arb funds are the close-end fund arbitrage funds that did CEF vs replicating ETFs or baskets or futures etc, perhaps less than 5 surviving funds with only 1 more managing than $100 mn. vol arb funds on the other hand have been very successful, although there are only a few of them. not sure about correlation trades - i know a few funds which lost big on those. a large category of funds are out there that do event driven trades (again not statistical), these are merger arb, stub trading, share class arb etc. many of them have been moderate to widely successful.

The vast majority of successful pure quantitative stat arb funds are not hedge funds in this arena. It's mostly prop shops that have the infrastructure and speed to play ultra high frequency stat arb game. They are not open for investments as a result they don't publish their returns or don't hear about them unless you are in the same circles. They are extremely successful in terms of their risk adjusted returns but they also do market making as well as pure arb trades. Some of these shops' good portion of their success comes from market making and speed related microstructural games as a result it may be hard to know how good they are at pure quantitative statarb.

I made a video tutorial on Pairs trading using Matlab feel free to check it out at http://tradingwithmatlab.com/pairstrading.htmlLet me know if you have any ?'s(domenic@tradingwithmatlab.com)

QuoteOriginally posted by: cmphilPair Trading typically requires purchasing very expensive software, but not anymore!I run a site which provides a FREE tool to make dollar-neutral pair trades. It also outputs a comprehensive data sheet that is very helpful when making trading decisions.An example of the output generated by the tool can be seen here: http://catalystcorner1.com/VZ_T.pdfHere is a link to the tool: http://catalystcorner.com/index.php?m=pair_toolYes everyone loves your spam all over any thread with Pari trading in the title...

Sorry - it just seems there are quite a few threads related to this topic and I thought everyone would be interested

QuoteOriginally posted by: cmphilSorry - it just seems there are quite a few threads related to this topic and I thought everyone would be interestedYou obviously did not do your homework Still, I grant that you did not post on Software, Numerical or "Off"