• The others train you to forecast, but our scholars train you to forecast accurately in the capital market

Spyros Skouras is an Associate Professor of International Finance at the Athens University of Economics & Business and a founder of Scientific Investments, a firm that applies cutting edge scientific research to analyze quant strategies and funds. In the past, he has held academic positions at Cambridge University, the Santa Fe Institute and Imperial College and has consulted for several leading quant funds. In addition, he contributed to the UK Secretary of State “Future of Computer Trading in Financial Markets Project” and co-authored regulatory impact assessments related to MIFID-2. Professor Skouras has a long-standing research interest in integrating econometric methods with technical analysis.

As early as 2001, he published “Financial Returns and Efficiency as seen by an Artificial Technical Analyst” in the Journal of Economic Dynamics and Control and “Learning to profit with discrete investment rules”, in Quantitative Finance. In addition, he has a broad interest in empirical finance and has published in several other leading academic journals such as the Journal of Econometrics and the Journal of Urban Economics. Professor Skouras is a graduate of Cambridge University, Universitat Pompeu Fabra and the European University Institute.

One of the of big areas of confusion in the marketplace is regarding terminology, or the distinction between high frequency trading, flash trading and algorithmic trading, which are all terms that seem to be interspersed but they do mean different things. Algorithmic trading is any trading system which is entirely automated, requiring no human decision beyond starting the algorithm. HFT is algorithmic trading which is speed sensitive, but what people mean by speed sensitivity varies (at the very extreme, strategy performance can be sensitive to a few hundred nanoseconds of latency, but sensitivity in the range of tens of microseconds to one second is more often relevant). Flash trading refers to a controversial order-type which is available to all investors on some venues, but effectively is only usable by HFT investors. It is controversial, because it clearly caters to just this specific segment of the investor community and some have argued that flash trading is harmful for market quality and other market participants (but this argument has not been supported by much empirical evidence).

How do we compare the financial returns between algorithmic vs human trading on the stock market? I do not hesitate to say that when judged on a risk adjusted basis, the very best algorithmic trading firms easily beat any discretionary trading firm in history, hands down. The classic example is the Medallion Fund, which according to the recent Bloomberg article has produced around 80% annualized returns (gross of fees) over almost three decades, delivering $55 billion to its investors with negligible draw-downs at any point throughout this period.

How do you connect “Ecology” to Al go-trading? Inefficiencies in financial markets drive traders to develop strategies that will exploit them, including Algo-trading strategies, and this process is what limits how inefficient a market can be. This process is extremely competitive, making inefficiencies diffuse and hard to detect and therefore trading strategies tends to be diverse and highly specialized. Strategies interact, because a strategy could take away inefficiencies, that another strategy is exploiting (e.g. mean-reversion strategies may eliminate trends), or a strategy creating entirely new opportunities (e.g. market making strategies may reduce spreads, making otherwise unprofitable arbitrages viable) or  a strategy reinforcing the patterns that it exploits (as may be the case with certain trend-following systems). It is very natural to think about this as an ecology and apply ecological principles distilled from the study of other ecosystems to measure, study, design and regulate the markets. Unfortunately, despite the work of a small group of researchers including myself, the value of this perspective remains underappreciated.

My topic and the reason behind my keynote speech in the coming up Milan Conference of IFTA, will focus on the distinction between technical analysis and algorithmic or quant trading and the key insights technical analysis has brought to the quant community. I will review the quant landscape focusing especially on strategies pursued widely by hedge funds, and I will discuss the current and future position of technical analysis in this landscape. Finally, I will discuss whether lukewarm performance experienced by most managers in this area of practice in the last two years is consistent with expected performance randomness or should alter our expectations moving forward. I’m delighted to be delivering this talk most of all as an acknowledgment of the role technical analysis played in shaping my early thinking about markets. More broadly, it is an opportunity to emphasize the vastly underappreciated contribution of technical analysis to the algorithmic trading revolution.

 

The article is produced from the office of Spyros Skouras under the auspices of IFTA, edited by  E. Tweneboah Senzu and Published at the Office of Bastiat Institute Ghana

 

 

 

Comments are closed.