The Future of Algorithmic Trading

Human beings – An Emotional Bunch!

We know that human beings are not always rational when making investment decisions. Have you ever overpaid for something in the heat of the moment? What happened? Well, it’s likely that your decision to purchase was driven by emotion rather than rational reasons. I’ve personally experienced many such instances, but one in particular stands out. Some time back I bought a piece of property in a stunning sea side village. Upon viewing the property I succumbed to a flood of emotions and as a result discounted the over-inflated price I would soon pay. My experience was a stressful one that emphasised the shortcomings of making emotional decisions. This story is all too common, and emotion is a very real phenomenon.

Emotion – Here to stay?

I don’t believe we can ever free ourselves entirely of emotion. In my mind, claiming full control of emotion is analogous to claiming control of one’s heartbeat. Emotion is hardwired into the essence our very being. Over the millennia evolution has moulded our psyche to respond to the environment in very specific ways. We quickly learned that it paid to be fearful in the face of a hungry lion (for those that didn’t, our gene pool only improved). Emotions like fear and greed have played a pivotal role in our success as a species, and they’re not going anywhere soon.

The Epitome of Emotional Decision Making

The true life story of Billy Beane, depicted in the brilliant movie Moneyball, is a great example of the effects of emotion. Beane, who managed a baseball team on a very small budget, came to the realisation that he could not build his team with the same approach as the big budget teams because he simply could not afford the best players. His solution was to instead employ statistics to uncover players that were tremendously undervalued due to the subjective biases of team scouts. For instance, one particular player pitched the ball with a peculiar style and as a result went overlooked. Others weren’t tall enough, good looking enough or had the right personality. However, this had nothing to do with how good they were as players; but they were consistently overlooked by big league teams.

Through the use of statistics Beane cut straight through the subjective biases of team scouts and assembled a team with less than half the budget of bigger teams. He subsequently went on to set a record of consecutive wins, changing the game of baseball forever. Incidentally, it’s no surprise that the Boston Red Sox, who is owned by former quantitative hedge fund manager, adopted Beane’s approach and went on to win the world cup.

The Solution to Emotion

Although emotion served us well to survive and succeed as a species, it does little to help within the realm of investment management. We tend to behave inappropriately when confronted with financial decisions under duress. The effects of human emotion on financial decisions have been well documented. An entire field of study, known as behavioural finance/economics, is dedicated to exploring these effects. The concepts are broadly accepted by the academic community. Emotions like greed, fear, regret, remorse, pride and ego play havoc with our ability to invest rationally and successfully. Most epic failures in business can be linked back to some form of emotion. So how do we control the risk of emotion in our trading?

One approach, which we use at QuantLab, is to employ a mechanical trading system that relies on a computer to generate objective trading signals by following an algorithm, or set of rules that determine when to enter and exit. With the advent of technology it’s become easier than ever to mechanise a process. In my view this is the simplest and most robust approach to calming the negative effects of emotion. But are there risks to a purely mechanical approach?

Algorithms – An Overcrowded Trade?

I’ve now seen many asset managers make reference in their marketing to objective rational processes as key drivers in their performance. It would appear that algorithms are slowly taking over the world, which leads one to ask: what are the implications of the apparent reliance on mechanical models? If everybody resorted to using an algorithm to trade then the market would slowly become more efficient and the opportunity to profit greatly diminished.

The truth, however, is quite different. Despite much press around algorithms and quantitative hedge funds, it’s still very difficult to sell a purely mechanical approach in the asset management world. Why? In walks emotion again. It turns out that there is a bias that human’s exhibit called the Algorithm Aversion bias.

Algorithm Aversion – Why Algorithms will Survive

I found the recent paper by by Dietvorst, Simmons and Massey which discusses the Algorithm Aversion bias in detail very interesting. It appears that humans are susceptible to cognitive biases even in the face of a well thought out objective model. The result is few traders can follow a mechanical approach with the kind of discipline required to extract all the inefficiency from the market. And because of that, discipined mechanical traders will continue to enjoy an edge in the market place. If you’re a mechanical trader, it’s important to understand this bias so that you don’t make the same mistakes.

Here’s the abstract:

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.

And here’s the full paper for those interested in reading it Algorithm Aversion

I hope you enjoyed this post. I always welcome your thoughts and comments. Feel free to send me an email.

Happy Trading,

Passionate about generating and sharing quantified trading models that empower individuals to trade successfully. I founded to realise my passion.

Leave a reply:

Site Footer