AQR has a paper out that looks at sizing liquid and illiquid alternatives into a diversified portfolio. Obviously there a lot of math involved and quite a bit of the paper devoted the role return assumptions play in their process. Keeping things simpler for this blog post, here's a graphic from the paper that cuts to the heart of the paper.
Since I'm not a fan of illiquid, aka private, we'll just stick with the first two columns. Yes their Long Term has some private equity but I think we work around that without creating a huge distortion while sticking with names we typically use for blogging purposes.
For the Long Term model I built out the following
I'm subbing CBOE and NOC in as proxies for private equity. Private equity ETFs own operating companies not private equity portfolios. ETFs that track the space are more volatile than the broad index without adding outperformance and if we go with Blackstone or KKR, which we've used in other blog posts, the results would be skewed to the point of being useless. CBOE and NOC are fudging a bit but we've used them in many previous blog posts and they've been client holdings for 11 and 20 years respectively.
And for Yield Based
Here's how it all sorts out.
I'm not sure what timeframe the AQR is using to get return numbers in the sixes versus what we have above in our backtest. The standard volatility numbers are similar but the Sharpe Ratios are much higher in our back test. Maybe the AQR table is a shorter period, giving more weight to 2022 but either way, the point is to take their idea to see if it can be made into a workable portfolio.
Looking at the year by year, 2020 was the only year where the two AQR portfolios got left way behind. The equity betas of AQR Long Term and AQR Yield Based are low at 0.62 and 0.78 respectively. The Calmar Ratios are much better but the Kurtosis for each one is inferior.
The first observation is the success of the two portfolios in 2022 despite the lack of any first responder alternatives, relying heavily on managed futures as a second responder defensive. If someone wanted to mimic this in real life, they could build any sort of equity portfolio with that sleeve, not just have one fund and it would make sense to use two or three managed futures funds like maybe one replication fund and one full strategy, same with splitting up the fixed income but I would avoid or greatly underweight any sort of duration. QLEIX is a very good fund but it would take a lot of faith to ever go 15% let alone 30% which I can't see myself wanting to do.
I'm not sure if what we've done in this post would constitute mimicking the AQR study or more like influenced by it. If influenced, that ties in better with the idea of taking bits of process from others to create your own process.
One note about all of theses posts, following up on a point I touched on earlier which is that I try to use the same names for these back tests over and over on purpose. I think the continuity is helpful for following the arc of the work we're doing and I think it is a nod to simplicity. I try to avoid using client holdings where possible which is easy where managed futures are concerned but less so client and personal holding BTAL, there no other fund like BTAL that I am aware of.
The information, analysis and opinions expressed herein reflect our judgment and opinions as of the date of writing and are subject to change at any time without notice. They are not intended to constitute legal, tax, securities or investment advice or a recommended course of action in any given situation.
No comments:
Post a Comment