Hedged Convexity Capture

[QZW 6]

I became interested in this strategy 

http://seekingalpha.com/article/2117623-part-vi-non-correlated-hedged-convexity-capture-revisited

The author that claims to do his own research, promotes a rather simple strategy, such as

 

I. Short SPXU with 40% of the dollar value of the portfolio.

II. Short TMV with 60% of the dollar value of the portfolio.

III. Rebalance weekly to maintain the 40%/60% dollar value weighting between the two instruments.

 

Looks pretty neat and simple. One of my former students who is a WS quant these days says that the charts in this link are probably correct. Looking at underlying, SPXU volume is decent, but TMV is rather small. I wonder if one can get jammed by poor availability of the shares to short. Any thoughts?

Thanks.

QZW 

 

[SeanM 29]

I started reading through these articles a few days ago, actually, but haven't had nearly enough time to get through and digest everything.  The one concern I noted right off the bat was that all historical performance / back-testing data I saw (again, I haven't read the entire thing, so I could be off with this comment) seemed to start in the early stages of the current bull market.  Was there anything about how this strategy fared during 2008?  2002?  I'm definitely interested in reading more on this strategy, though the author's brashness and dismissal of all other investment strategies makes me roll my eyes.

[QZW 6]

I started reading through these articles a few days ago, actually, but haven't had nearly enough time to get through and digest everything.  The one concern I noted right off the bat was that all historical performance / back-testing data I saw (again, I haven't read the entire thing, so I could be off with this comment) seemed to start in the early stages of the current bull market.  Was there anything about how this strategy fared during 2008?  2002?  I'm definitely interested in reading more on this strategy, though the author's brashness and dismissal of all other investment strategies makes me roll my eyes.

I am not sure if the leveraged names that the author pushes through the blogosphere were available in 2008 or prior dates. He is brash, I agree. But bravado beside, the long inverse combo strategies that he presents may make sense. With the short/short strategies, I don't know. I am not sure spending $200 for his book is justified, but the strategy probably is worth a thought. I wonder if Jesse and Chris who have experience with long/short leveraged ETFs want to share any input.

Edited March 31, 2014 by QZW

[dbh21 62]

I tend not to believe anything until I test it myself, but this is quite simple to test. Definitely going to when I get a few moments. Should be pretty trivial to reproduce. Love the simplicity.

 

Looks like 2011 might have to be an example of how it would fair in a bad market.

 

I'll post results if I get around to it.

[dbh21 62]

Crikey... $200 for a 26 page book. Guess its not going to be in the Kindle Lenders library  

[QZW 6]

I tend not to believe anything until I test it myself, but this is quite simple to test. Definitely going to when I get a few moments. Should be pretty trivial to reproduce. Love the simplicity.

 

Looks like 2011 might have to be an example of how it would fair in a bad market.

 

I'll post results if I get around to it.

Thanks, dbh21. I will be looking forward to your post. The days when I could test any strategy myself are over. Indeed, 2011 sounds like a good example of a bad market. 2008 is an aberration.

QZW

[dbh21 62]

I've done some preliminary work. My method was simple. I used the adjusted closes on a weekly basis using data from yahoo. No compounding. No reinvestment of dividends. See the spreadsheet attached.

 

I'm getting good results, but different than Harry's. I'm in the same ballpark, but my values are better for some reason. I can't figure out why.

I notice he uses etfreplay.com, so it could have something to do with its data maybe, or more likely, I have a bug somewhere. I did not take into account short margin rates, so maybe that is a part of it.

 

Data only goes back to 2009, so no data prior to that. You can also see with a modest portfolio you can end up shorting 10% of the daily volume of TMV pretty quickly. Not sure how easy it is to short or not.

 

 

I also noticed this little gem

Investors considering a Non-Correlated Hedged Convexity Capture Strategy make a huge mistake when they believe that such a strategy depends upon bull markets in equities or bonds. While I totally concede that discontinuous drops (flash crashes, surprise crises, etc) in markets can have devastating effects on a Hedged Convexity Capture Strategy (which is why the non-public version we have created does not use shorting), the notion that secular bull markets in equities and bonds are required for the strategy to enjoy a positive return is incorrect.

 

 

My first guess is he uses DITM puts or even a synthetic short position, or a backspread.

 

Non-Correlated Hedged Convexity Capture.xlsx

[dbh21 62]

Because of the low liquidy of TMV and SPUX, last night I looked at TBT and SDS as alternatives as they appear to be much more liquid. But their returns were about 100% less (350% vs 450%) over the 5 years worth of data. Still good returns, but lower. I tested a few other options. Adding TVIX or VXX adds a little to the performance with more volatility (ha...). The author appears to have picked the best instruments from what I have seen so far.

[dbh21 62]

I was thinking more about this strategy last night. It does well on sideways or bull markets. But shouldn't it get absolutely hammered in bear markets? Its long the market, 60 short an ultra long, and 40 short ultra-short bonds (which should go up in a bear if the world worked as expected).

[QZW 6]

Yes, looking at the strategy, it is best to start implementing at the market bottom (like anything else, I suppose, except for straight shorting). There will be losses on the way down. Besides, we are looking at borrowing costs of about 3% annually. So I think it is a feasible thing, but I put it on hold. Nothing beats buy low/sell high strategy.

Edited April 10, 2014 by QZW

[Schmu 0]

I think the major risk to this strategy is in a rising interest rates scenario where neither treasuries or stocks are doing very well. I've played around with adding a small percentage of short XIV to the mix at the expense of some of the treasuries but it's hard to tell how well this would do when the last time rates went up was before most of these securities even existed.