• Milton Friedman Society

Quantamental Investing

Quantamental investing is a new investment strategy combining fundamental evaluation performed by analysts, which is based on subjective “qualitative” parameters, with quantitative data gathered by algorithms and AI systems. This approach aspires to improve the stock picking process by blurring the line between man and machine, capturing the best of both worlds.

Both fundamental and quantitative investing aim to generate as much alpha as possible by beating the market; however, the strategies implemented to achieve this goal are radically different. Fundamental investors are highly qualified professionals who try to assess a company’s intrinsic value by scouring through countless financial statements, determining whether the firm is undervalued or overvalued compared to the market price. Quantitative investors use instead advanced mathematical and statistical algorithms to quickly analyze data and exploit patterns arising from market dynamics.

Fundamental investing suffers from some of the same issues affecting human behavior, such as unconscious biases towards certain stocks; for this reason, an objective analysis like the quantitative approach can seem more appealing to portfolio managers. However, this strategy has its own drawbacks, for example, to reduce the risk arising from a single stock, the composition of the portfolio needs to be changed quite frequently, increasing trading costs. Including fundamental factors in the portfolio creation process allows for longer term positioning with less need for security turnaround and expensive transactions fees.

No human being can read thousands of earnings reports or listen to more than a couple of conference calls, but on the flipside, machines are not always able to put these results into a context and infer the right result.

A good example of the two approaches coming together is the evolution of “Alpha capture”, a quant trading strategy which scouts for any kind of investment banks’ recommendations and comments and develops them into buy/sell signals. As this technique became more widespread in the 2000s, the profit margins decreased substantially, signaling the need for an improved approach to the Alpha capture strategy. Among the first hedge funds to respond to the changing market conditions was GSA Capital, which in 2015 hired a couple of outstanding discretionary portfolio managers in order to implement their views in the quant algorithm. The initial results have been promising, demonstrated by GSA’s CEO comments: “Bringing together the power of machines and human brains is a cliché, but it is the future of investing”.

Another example comes from the development of machine learning systems; while these programs uncover patterns by analyzing enormous and varied data sets, from words and numbers to satellite images and sounds, it does not mean that every pattern can be modelled into a profitable investing strategy. Portfolio managers need to combine their own views on future market trends with the relevant patterns coming from machines, keeping in mind the possibility of major shocks such as the 2008 financial crisis, which could catch automatic models blind sighted and potentially lead to huge losses.

A report by Morgan Stanley’s investment management team published in 2018 claims that about 65% of global equity manager’s excess returns are driven by quantitative analysis. This means that the remaining 35% is up for grabs, and can be traced back to a portfolio manager’s stock picking ability. Some investors might decide to simply increase the number of securities in order to reduce market risk thorough diversification, leaving as the only source of risk (and hence return) the timing of the investments, all the while increasing trading expenses.

However, portfolio managers can improve the quantitative framework by taking informed decisions on individual stocks, potentially delivering even higher excess returns. By combining these two approaches, the alpha of a portfolio will effectively depend on two streams of returns (instead of one); one on the market level, which is handled by the quantitative tools, and the other on the individual stock level, chosen instead by the manager.

This strategy can generate alpha more consistently over time, as in periods of market downturn the stocks chosen through fundamental analysis might perform well, offsetting the losses deriving from the market exposure of the quantitative strategy.

The report shows how more consistent returns greatly increase the benefits of compounding. Furthermore, investors tend to prefer less volatile performance, keeping them from getting caught up in the market’s exuberance and taking any irrational decisions.

Many hedge funds are approaching quantamental investing, but DE Shaw is by far one of the most fascinating and successful cases. Employing a mix of quantitative and discretionary strategies to manage its over $50 billion worth of assets, DE Shaw has established itself as an important player in the industry.

This strategy has evolved from the fund’s early days in the 1980s, when it implemented pure quant trading strategies; this change left some executives dissatisfied with the new approach, leading them to create their own quant hedge fund, Two Sigma.

While the exact trading strategy is shrouded in secrecy, the stellar results are undisputable; according to LCH Investments, DE Shaw has made over $29 billion for its clients since its inception, outperforming its benchmark even in periods of dire market conditions.

The quantamental procedure is demonstrated by the employees of the fund, consisting mostly of computer scientists but including also 25 international math Olympiad medal holders and important political and financial figures, the likes of Larry Summers, former US Treasury secretary.

The use of discretionary investing by the fund was evident when it swiftly responded to the Crimean dispute between Russia and Ukraine in 2014 and the Volkswagen emission scandal in 2015; algorithms are not able to deal with unexpected events which shock financial markets, signaling the need for human intervention. Another instance of discretionary investing by the fund occurs when trying to exploit the differences in price performance between Tencent and Naspers, which owns one third of the Chinese conglomerate; while quantitative signals are used, the portfolio manager’s view on the different market and political conditions in the two countries is essential to assess the investment decision.

While its use is becoming more widespread, quantamental investing remains a controversial issue drawing ire by both sides of the debate; quants believe that traditional portfolio managers will never be able to fully comprehend the extent of quant strategies, old-school investors on the other hand complain that this phenomenon is shifting the focus overwhelmingly towards short-term results, losing sight of the long-term performance of a fund.

This strategy is likely to continue its expansion, in spite of developments in algorithms and AI, because as Ravid Mandell (JPMAM chief data scientist) puts it: “There’s stuff that happens in the human brain that is so hard to replicate.”


 

Bibliography


Canalyst . (2019). Canalyst newsletter: quantamentally bullish.

Constable, S. (2019). What is 'Quantamental' Investing? Retrieved from Wall Street Journal: https://www.wsj.com/articles/what-is-quantamental-investing-11554688800

LePan, N. (2019). Human Insight, Computer Power: What is Quantamental Investing? Retrieved from Visual Capitalist: https://www.visualcapitalist.com/what-is-quantamental-investing/

Morgan Stanley Investment Management. (2018). Quantamental Investing: the future is now.

Pozen, R. (2019). Will bots replace humans in active equity investment? Retrieved from Financial Times: https://www.ft.com/content/efe4f97a-adb1-3cea-b098-2b616b5ce531

Sueppel, R. (2019). How to build a quantamental system for investment management. Retrieved from Systemic Risk and Systematic Value: http://www.sr-sv.com/how-to-build-a-quantamental-system/

Walker, O. (2019). Money managers take cues from ‘Moneyball’ playbook. Retrieved from Financial Times: https://www.ft.com/content/42e5c41a-ee66-35df-a600-b6654d3152cb

Wigglesworh, R. (2019). Quants seek human touch in reboot of investing strategy. Retrieved from Financial Times: https://www.ft.com/content/0093dcd4-ad59-11e9-8030-530adfa879c2

Wiggleswoth, R. (2018). The rise of ‘quantamental’ investing: where man and machine meet. Retrieved from Financial Times.

Wiggleswoth, R. (2019). DE Shaw: inside Manhattan’s ‘Silicon Valley’ hedge fund. Retrieved from Financial Times: https://www.ft.com/content/0364850c-3ebf-11e9-9bee-efab61506f44

6 visualizzazioni0 commenti

Post recenti

Mostra tutti