Keeping in similar vein to a number of the previous entries about the mysteries of defining, selecting and tracking hedge fund strategies here is a 2015 paper by Swiss wealth manager Pictet titled “Hedge fund indices: how representative are they?”
Amongst its gems covering the biases of the indices is this great graphic explaining the source of ‘self-selection’ bias:
Core message: less than one percent of funds report to all the tracking databases. One study puts the impact on the performance figures of this at 1.9% per annum. More worryingly, the cumulative impact of all biases may be as high as 10.7% annually.
Over time that adds up to a lot of dispersion between index providers…
An earlier entry pointed out how many sub-strategies there are under the title ‘long/short’. Take one step back and consider this table from the 2016 Preqin Global Hedge Fund Report (click for a larger, legible version):
The table continues for an entire second page but this half makes the point: that’s a lot of headline strategies (before any talk of sub-strategies).
For those who like steady returns with as few shocks as possible ranking this list on the 5 year net return/volatility ratio produces a clear winner in the equity class: RV Equity Market Neutral (our yellow highlight). It is not the best on that ratio overall: two credit-based approaches formerly best known for cameos in The Big Short pipped it.
The top 10 strategies on the return/volatility ratio look like this:
For clients running our software it is a useful reference point when designing strategy approaches.
So consider value as an investment category. Words to the effect that “the market always recognises the economic fundamentals of sound enterprises in due course” are part of the strap lines of value managers: intrinsic value against market prices; 60 cents for a dollar; and so on.
In overvalued and irrationally exuberant markets this approach frequently provides poor returns relative to benchmarks as market participants suspend good sense. Still, after the fallout of such episodes the value manager’s reputation as a paragon of sober logic is burnished.
And thus the proposition that the oft mad market will come round and fully price the fundamentals of a value enterprise remains seductive. So much so that the possibility of the fundamentals coming round to align with prices is frequently overlooked.
Time does not heal all mispricings. Much can go wrong as it elapses – and as the architecture of financial markets evolves.
Last month, for example, Francis Chou was reported by Bloomberg as returning his 2015 advisory fee in an act of solidarity with his investors who lost 22% in his Opportunity Fund. Mr. Chou had a poor 2015 but, historically, is a strong value manager – as he points out in the article by referring to his great long-term record. Still, 22% is quite a drawdown to overcome in reasonable time without taking excessive risks.
An outlier? Some readers may be thinking “Warren Buffet” just about now. Below is a chart of Berkshire Hathaway’s performance versus the S&P 500 on a rolling 5 year basis.
Source: Berkshire Hathaway annual letter, 2015
Even the King of Value has found the job tougher and tougher since the turn of the century.
Long/short – what could go wrong? Hedged positions, protection etc etc etc.
And yet there are headlines like this one:
One of the problematic aspects of such a headline is that “long/short” is a hard category to define – and there is no consensus around the term. Which leaves plenty of room for whatever angle one wishes to shock and amaze with.
A cursory examination will show that Long/Short is a broad church – and one with some disciples who regularly trek over to the neighbouring tabernacle of Relative Value to worship.
Thus probably it is worth defining precisely what sub-class of strategy is being analyzed before daubing all with the same brush.
Two competing – but sometimes overlapping – approaches to pairs trading are to take either a fundamental ‘signal’ view based on changes to factors such as the accounts, the economy, the CFO’s penchant for recognizing revenue early and so on; and the ‘noise’ approach whereby the trader concentrates on the divergence in value of the instruments for reasons unrelated to changes in fundamental conditions.
Bloomberg occasionally publish pairs trading ideas in the first category. Like this one for HCN/SPG. One interpretation suggests that this is really a macro call using the pair as a proxy and the 10 year bond yield as a trigger.
You may wonder how to avoid over-reliance on those analysts forecasts and GDP predictions cited (see this from Larry Summers for why that might be a concern) when trying to apply this idea. Even the yield differential quoted as an advantage depends on how the trade is set up (money neutral vs beta neutral for example).
But what of history? Taking the 2001 and 2007 recessions as precedents, as the piece does, show the trade would have made 16% over 8 months and 35% over 18 months respectively (on a dollar neutral basis).
Not bad – but 2 trade entry data points is not a trend. And outside of those rather difficult-to-time periods SPG has strongly outperformed HCN by a wide margin since 2002 (this is shown on the Indexed Prices & Spread graph below).
The noise approach (here lend some rigour with cointegration) would have used far smaller holding times inside both those recessionary periods. Taking the August 2007 to March 2009 meltdown, for example, an uncomplicated strategy (backtest graph below) would have traded 4 times for a total gain of 17% (and an average return on capital invested per trade of 4.2%). The total holding period of 2 months means, once annualized, that ‘noise’ beat ‘signal’ by over 4 to 1.
Horses for courses – but the ideal approach tends to combine both.
Some visual analysis:
The back test results:
Not entirely satisfying but cointegration is a linear association between multiple time series and most prone to break when a shock occurs. Shocks (excluding calendar events such as earnings announcements) tend to be unheralded.
However, for entertainment value the question can be answered with a high degree of statistical confidence. This Google Book preview link to Professor John D. Barrow’s entertaining book “100 Essential Things You Didn’t Know You Didn’t Know” explains.
The moral, maybe, is that statistical confidence is not the only consideration to take account of in decision-making!
Running Forex pairs in ArbMaker differs from equities in critical ways, especially at the intraday level. Here, for example, are a few points worth considering very carefully:
- Spreads on many FX instruments can be significantly difficult to cover in a pairs trade – all the more so at higher frequency intervals like 5 and 15 minutes
- Cointegrated relationships change/break faster at higher frequency levels and require careful monitoring
- FX is particularly volatile due to near daily rafts of macro-economic announcements
- It is much harder to dilute risk in FX due to fewer available instruments and the dominant influence of the USD in the market
The typical use of easily available high leverage for FX trades amplifies the downside of all of the above. And on top of all this trading triagular FX pairs can make it all worse. That is to say pairs such as EURGBP/EURUSD where EUR appears on both sides of the trade.
Depending on the order of the EURs these ‘triangles’ either cancel out or double up a short/long position. Tricky to trade and, for the unwary, can lead to big holes in one’s trading capital.
If you do not want to risk them simply filter the triangles out of your Scheduler Job options like this:
At each update in the Tracker new data is added to the time series and its impact on the statistical profile integrated.
Depending on what that profile looks like over time users may want to discard certain pairs (for instance, if they are no longer cointegrated at 95% and so on – but the criteria used to keep/discard is entirely up to the trader).
In our internal intraday portfolios it is this process that tends to drive the frequency/sequence to run scans for pairs already in the Watchlist and/or Tracker. As we lose pairs to failing cointegration we will run new scans.
Why? The premise of the software is to:
- allow users to construct their own pairs trading strategies; and
- to ensure those strategies are applied to cointegrated pairs.
Note that the software is designed to control the latter point through deployment of our ‘Dynamic Filters‘ In the Tracker. These monitor the stat profile with each new piece of incoming data.
Trials of version 4.0 can be had here.
- The scan was for a small number of combinations. The smaller the number of pairs scanned the fewer the results. By default the ‘Coint. Scan’ screen returns only cointegrated pairs. So if it finds nothing there will be no returns. However, by ticking the ‘Bypass’ box on the ‘Coint Scan’ screen the software will return the fails also. This enables users to see at which point in the process certain pairs failed cointegration. There is a limit of 500 combinations using the Bypass but it will allow you to see what is going on with the lack of returns.
- The date range used uncovered no cointegrated pairs. Changing the date range will change the statistical profile of the scan returns (both the fails and any cointegrated pairs). Similarly, end-of-day and intraday scans on the same symbols will produce different results.
- The feed might not be configured correctly. This can be tested by going to Parameters->Symbols and clicking the ‘View’ button on one of your symbols. If the graph that appears is populated everything is OK. If not double check the configuration using the documentation instructions.