1.
지난 몇 년을 되돌아 보면 트레이딩을 하는 사람들중 API나 알고리즘을 우상화하는 경향이 있었습니다. 손매매는 후진적이고 손실을 볼 수 밖에 없고 API나 알고리즘을 이용한 기계매매만이 진보적이고 이익을 낼 수 있다는 생각입니다. 시스템트레이딩이나 API를 이용한 트레이딩을 설명하는 강연을 보면 이런 경향이 두드러집니다. 트레이딩의 목적이 수익이라고 하면 수익을 내는 방법에 정답은 없습니다. 그렇지만 어떤 방법이 투자에 더 도움을 주는지, 기계가 시장을 주도하는 시대에 주요한 화두중 하나입니다. 이를 표현한 문장이
“models beat experts”
입니다. 모델은 시스템에 의한 기계적인 의사결정을 말합니다. 알고리즘과 전문가중 어디가 좋은 성적을 내는지, 이를 다룬 많은 연구가 있었다고 합니다. 이의 결론은 모델입니다.
What about becoming an investment expert, buying trading or analysis software, and spending countless hours researching investments? While research may be entertaining. “countless hours of being busy” will not add value to your investment results. Even the professionals suffer from this faulty premise. A number of research studies have shown that there are far fewer money managers with statistically significant positive alphas (superior performance) than would be expected by random chance. And even then, other research demonstrated that what may appear to be statistically significant positive alpha doesn’t indicate skill.
The reality is that models (systematic decision-making) beat expert judgment. In fact, research has shown the phenomenon is happens of variety of fields. For example, a review paper examined the results of 136 research studies comparing statistical prediction methods to clinicians’ judgment in psychological assessments. The conclusion was that the statistical method was superior to expert judgment, regardless of the expert’s level of expertise. Models equaled or beat expert judgment 94% of the time.
Models or Experts – What Will Make You More Money? 중에서
그렇지만 현실은 다릅니다. 두가지 현상이 나타납니다. 정량적인 시스템의 수익률이 기대에 미치지 못하고, 이 때문에 예상을 빗나가는 경험으로 투자자들이 알고리즘을 거부하는 현상입니다. 두가지 현상을 분석한 논문을 소개합니다.
먼저 “왜 정량적인 시스템의 수익율이 기대에 미치지 못하는지’를 분석한 논문입니다. 제목은 Why Most Quantitative Investing and Trading Systems Fail입니다.
논문이 주장하는 원인은 ‘logical fallacy’입니다. 논리적 오류를 줄이는 방법은 다양한 경우를 전제로 모델링을 하는 것입니다. 저자는 Invert와 Sufficient로 표현하였습니다.
Avoiding the fallacy of the converse: Invert, always invert
The key issue here is that most quant systems seek out factors that were associated with trades/investments that generated superior investment returns. Such a process ignores Jacobi’s insight, “Invert, Always Invert.” It is as important, if not more so, to understand those cases that shared the same characteristics but did not work well.
“Invert, Always Invert.” – Carl Gustav Jacob Jacobi, German Mathematician
“Hundreds of studies have shown that wherever we have sufficient information to build a model, it will perform better than most people.” – Daniel Kahneman (as you read this statement, don’t forget to consider the implication of the word “sufficient”)
다음은 알고리즘을 부정하는 현상에 대한 분석입니다. Algorithm Aversion — Why people don’t follow the model!이 제목입니다. 눈문에 나온 사례입니다.
Research shows that evidence-based algorithms more accurately predict the future than do human forecasters. Yet when forecasters are deciding whether to use a human forecaster or a statistical algorithm, they often choose the human forecaster. This phenomenon, which we call algorithm aversion, is costly, and it is important to understand its causes. We show that people are especially averse to algorithmic forecasters after seeing them perform, even when they see them outperform a human forecaster. This is because people more quickly lose confidence in algorithmic than human forecasters after seeing them make the same mistake. In 5 studies, participants either saw an algorithm make forecasts, a human make forecasts, both, or neither. They then decided whether to tie their incentives to the future predictions of the algorithm or the human. Participants who saw the algorithm perform were less confident in it, and less likely to choose it over an inferior human forecaster. This was true even among those who saw the algorithm outperform the human.
영화 제목에 빗대면 투자자는 ‘참을 수 없는 존재의 가벼움’때문에 쉽게 알고리즘을 부정합니다.
알고리즘의 우위를 믿는 트레이더라고 하면 어떻게 해야 할까요? 어떻게 투자자에게 설명해야 할까요? 위의 논문을 소개한 Alphaarchitect의 의견은 아래와 같습니다.
Successful investors understand that models will fail at times; however, being able to stick with the model through thick and thin is a good strategy for long-term wealth creation.
2.
앞서 Model beats a expert를 소개했습니다. 이와 직접적인 관계는 없지만 시장의 변화가 인간의 심리에 미치는 영향을 분석한 논문이 있습니다. 가격의 급격한 변동(하락)을 사람에게 위험을 회피하려는 행동을 하게끔 한다고 합니다. 비록 수익이 낼 가능성이 높더라도. 이 때문에 모델이 필요한 것이 아닐까 합니다.
Consider the concept of dynamic risk aversion, which is the idea that human beings don’t stick to a set risk/reward behavior—their appetite for risk can change depending on their recent experience… As market prices drop below the twenty percent threshold, an economist assumes that the new price is a bargain. Expected returns have gone up after prices have moved down, while volatility and risk aversion are assumed to be relatively constant…
But this doesn’t happen. Stocks can—and have—gone down over fifty percent, and these movements are much more volatile than the underlying dividends and cash flows of the stocks they represent! Remember 2008/2009? How many investors swooped in to buy value versus threw the baby out with the bathwater and kept selling? …
One approach to understanding this puzzle is by challenging the assumption that investors maintain a constant aversion to risk. Consider the possibility that investors change their view on risk after a steep drawdown (i.e., they just lived through an earthquake). Even though expected returns go up dramatically, risk aversion shoots up dramatically as well.
How Market Volatility Affects Our Brains중에서
All Trading is Quant Trading을 보면 재미있는 표현이 나옵니다. 항상 R&D를 해야 하는 이유로 읽었습니다.
The good thing about finance is that something that sounds heretical today can become ubiquitous in the future, especially if it works.
