Trading has been around for a while and even though the process around it has evolved over the years, the fundamentals are still the same. We buy and sell securities or other financial instruments in the market in a hope to make money. The positive difference resulting from the trade is our profit while the negative difference is our loss. However, taking a deep dive into this process gives us an idea of how sophisticated managing successful trading strategies can be.
A rookie trader might not have all the knowledge necessary to enter the market with a profitable strategy right off the bat however, it has been argued that trading strategies can be taught by the famous Turtle Experiment. While it might seem like a good idea to invest in an upcoming IPO, without properly calculating the risk associated with it could lead to a disaster. History has shown many big organizations have been brought down to their knees simply because they couldn’t see the risks associated with their investments and their trading strategies and decided not to alter them with changes in the market. A lot of small players lose in the trading market because they have a get-rich-quick mentality and become the victim of behavioral market manipulations. They absorb the fallacy based on their own bias and might blindside an obvious signal. For example, if someone is a huge admirer of Apple, they might have a strong bias towards Apple stock thinking that its stock is going to skyrocket simply because they love Apple products and it has always delivered for them. However, that person might fall flat with a hot-hand-fallacy and lose money if he has had trends of past success with Apple stocks, however, he/she is reluctant to change their positions even if the market signals the stock going down.
Larger firms often use hedge funds to create a systematic trading strategy which minimizes the risk. They diversify their portfolio in such a way that, if one signal is bearish, they keep the risk under control by either shorting them or buying securities that are bullish. This is just an example of one of the many strategies big hedge fund companies use to give their client a competitive advantage. These strategies combined with a loose regulation and secrecy with their rigorously tested proprietary models give their clients a sense of satisfaction. These kinds of strategies help them be ahead of others in risk management and hence they are highly successful.
With the advancement in machine learning and artificial intelligence, algorithmic trading has gained a lot of traction these days. This adds a lot of opportunity for bigger hedge funds as well as a single day-trader operating from a basement. One of the various opportunities that algorithmic trading brings is the ability to run various simulations with different signals that are in the market. These kinds of tests emulated in various scenarios can give a good yield in an investment. Availability of cheap compute resources as well as trading API’s combined with a well built and backtested systematic trading model will even give small players a competitive advantage over their rookie players.
One thing we need to keep in mind is that machines don’t have emotions (yet). Hence depending on the type of a programmer or data scientists, we should be very careful about machine bias or in more technical terms ‘algorithmic-bias’. This topic often gets lost in many buzzwords that are in the industry today however, it is important to understand the challenges this brings to a trillion dollar industry. How would someone explain that the proprietary algorithms that have been used in a machine learning model to facilitate systematic trading don’t favor the corporation that developed it? Just as human cognition functions and gets into an infinite loop of fallacy in an abyss based on individual bias, machines are no different. Machines look and work merely on data sets that are fed to them to come up with predictive algorithms, however, it is challenging to identify if the data sets that are building these predictive models are not biased.