‘Quantamental’ solution to CTA underperformance proposed

Related Content Related Articles

Systematic investing, in the form of managed futures, often called CTAs, rose to the attention of the wider investment community particularly through its performance in 2008, when it delivered strong returns against the dramatic losses posted by most other strategies. In the following years, however, its performance has been somewhat lackluster, prompting several market commentators to wonder if the managed futures strategy was ever to come back and whether the diversifying performance of 2008 was something of the past, never to be seen again.

All strategies go through periods of underperformance and managed futures is no exception. The long grind however did open up potential for some managers to evolve their systems to better adapt to unusual conditions in future.

One type of evolutionary model that has developed is a combination of traditional Systematic investing with fundamental macro. This strategy aims to take the positive attributes of systematic investing, particularly those that make the strategy so powerful, whilst reducing the impact of its shortcomings. This solution preserves the power of systematic investing while addressing the drawbacks.

The positive attributes of systematic investing

Systematic investing has the ability to neutralise the negative impact of emotions, fatigue and cognitive biases within human decision making. Systems, by design, entail the consistent execution of a set of rules and are hence immune from the natural vagaries of the human brain.

The approach allows the benefit of seeing how an idea would have performed in the past if consistently applied. Simulations got man to the moon. We build cars, airplanes and explore into the outer realms of space through testing in the ‘lab’ for the unknown. Unusual environments can be synthesized and simulated to test out ideas; the ideas in turn can be stress tested to pull and stretch our imaginations.

Therefore, it has the power to efficiently process huge volumes of data, delivering a double-advantage: maximising the information content which can be extracted from the analysed data, as well as providing the broadest possible opportunity set of assets to select from.

Through a systematic approach we can also significantly reduce a company’s reliance on key individuals. No matter how influential a key person is in the development of a system, the knowledge is incorporated in the model which survives as an asset of the firm through its intellectual property. This can be invaluable.

Drawbacks of systematic investing

As with all investment strategies, a traditional systematic approach comes with some intrinsic weaknesses.

It is important to note the dependency on past data, and therefore the inherently backward-looking nature of the systematic approach. True, techniques have been developed to minimise the drawbacks from this, but it is an easy temptation to fit the approach to conditions of the past, believing they will repeat. Sheer speed of computational power increases that temptation of data mining.

As a result of the above, there results with the benefit of hindsight a self-perpetuating practice of changing parameters or models frequently, often known as optimisation. The ‘quant’ nature of the approach exposes it to the risk of ‘over-engineering’, such as a proliferation of variables, dependence on too many signals, constant tweaking of parameters. The negative outcome is a model that is overly back-fitted to past conditions, with the risk of decaying quickly or, even worse, being suitable only for the specific conditions it was developed in.

There is often an inability in this approach to put an economic or financial connection to the data analysed. The system by design typically treats the data in an aseptic manner, overlooking the conditions which generated such data in the first place. This ‘black box’ approach leads to a loss of intuition in system behaviour dealing with certain market environments.

In systematic investing often times different models are used to deal with specific environments. Subjectivity hovers around how much capital to allocate to say trend-following, mean-reversion, pattern-recognition, or other strategies.

The above weaknesses in the systematic approach lead to an inability to look forward and prepare for sudden, and frequently dramatic, changes in market paradigms. These have been most visible in the economic environment post-2008 crisis, with the disappointing performance of the strategy for an extended period of time. The exceptional intervention of financial and monetary authorities to avert a global meltdown has essentially overridden market fundamentals in this period. The consequence has been trendless, choppy, risk-on/risk-off markets, depressed volatility and spiked up correlations between assets. These are all unfavourable conditions for systematic investing.

preloader
Close Window
View the Magazine





You need to fill all required fields!