Blog

Machine Learning for Stock Price Prediction

Machine Learning for Stock Price Prediction

It’s no surprise that traders have been relying on tech when making investment decisions for a long time. Computers have been long employed for their outstanding capacity to make thousands of trades in a second. In this sphere, typical quant software becomes a thing of the past with AI-based software solutions gaining more recognition by traders and stockbrokers.

Some of the highly-reputable hedge funds and investment companies already use Artificial intelligence and Machine Learning in trading. Such companies as Renaissance Technologies, Hudson River Trading company, Two Sigma, Bridgewater Associates have been consistently successful in their automated trading strategies.

As many AI-based systems helped its adopters generate very high returns, many investors and traders are turning their eye on the technology. It’s definitely the time to think over AI place in modern and future trading. Read on to learn more.

 

AI strategies Outperform Human Traders and Quant software

Machine Learning hedge funds have outperformed human traders, average global hedge funds and quant software in past years:

Source: Eurekahedge

AI/Machine Learning Hedge Fund Index vs. Traditional hedge fund Indexes

Here comes the logical question – why? Here are the short answers:

Crunching Loads of Data at High Speed

In trading, every second could cost a fortune – for you or your clients. There is no time to blink –  anything can signal that a stock will rise or fall. And to make a right bet, these signals must be revealed immediately.

Without question, processing speed of a computer is incomparable with that of a human.

AI software scans financial reports and press releases for keywords, analyses similar price movement patterns from the past, and comes up with the suggestion – all that in a couple of seconds.

Predictive power improves over time

Rich volume of quality data plus well-thought-out machine learning algorithms are the basis for a successful AI-software. But it can go even further. As the time passes, ML systems learn on their own, without being explicitly programmed, therefore enhancing its accuracy of predictions.

Also, machines are devoid of any emotion, unlike a human being. Needless to say that there is no room for emotions when making trading and investment decisions.

AI-driven hedge funds

Scarce amount of performance data for AI strategies could be found, given such software’s proprietary nature. In fact, no one knows how the AI is leveraged exactly except for the companies themselves, as such AI software is always proprietary. Also, note that it would be meaningless to reveal the software’s concepts – then others would build the same, thus making trades unprofitable.

But still, we’ve found some interesting facts.

 

1. Informative Statistics from Eurekahedge

AI-driven hedge funds around the world appear to be doing pretty well. Eurekahedge, a hedge fund research firm, provides performance of indexes in numbers as well.  Here is the table from one of their report:

Source: Eurekahedge

Performance in numbers – AI/ML Hedge Fund Index vs. Quants and Traditional hedge funds

 

2. The secret of Trading Success from an AI-utilizing Trading Guru

Yoshinori Nomura, a director at the Simplex Asset Management, spent over 3 years developing AI-based trading software.

It has already spotted wise investments. One of the most notable examples occurred in 2016 (the case is precisely described in the Bloomberg article). Nomura’s AI-based program sold off Japanese futures, which seemed to be the wrong bet. But soon Britain proclaimed it’s decision to exit the EU, which triggered a major stock price drop. Simplex ended that day with its best result in 3 months, whereas many had lost out.

The system makes decisions on whether to buy or sell futures twice a day. It was trusted to move up to as much as a half of the fund’s value at any moment. In 2016,  the system was reported to oversee not less than 3.5 billion yen, which was equivalent to $34.9 million.

One of the biggest advantages is that the system it’s able to adapt well to constantly changing market conditions: it combines ML and quantitative data, considers market environment, market momentum and mean reversion theory at the same time. Y. Nomura, reveals his secret: “Among several keys to success is a system that monitors the market environment and adapts strategy to best fit current conditions”.

 

3. A story about integrating an AI-based trading system

At Man Group, the third-largest hedge fund by assets under management, AI was initially looked upon with skepticism. But now they make it their cornerstone strategy. Company is investing considerably in computer equipment and highly-qualified computer engineers to ensure its technological growth.

The company was among the first to apply AI in trading. Though, it didn’t trust AI technology to manage its assets from the start. Moreover, because of working principles of AI no one, including engineers, can explain why AI software makes the decisions it makes. And when there’s talk of money and investment decisions, investors wouldn’t take the absence of explanation on this.

After a series of testing efforts, the company experimented with their AI software by trusting it to handle some sum of money from a managed portfolio. Then, they handed to it then a bit more and more. The system proved to be profitable. The firm became comfortable with the technology, and more importantly – confident with it. By 2015, the company’s AI-based system was contributing roughly half the profits in one of Man’s biggest funds.

Four Man funds incorporating AI manage $12.3 billion in total. AHL Dimension fund assets have more than quintupled, meanwhile company’s assets under management have increased by 77% since the beginning of 2014:

Source:a Bloomberg article

 

Describing an ML-based system for Trading

Machine Learning is much about prediction. It has already been applied to predict future customer behavior and proved to be successful.

And can it predict whether the price on stocks would go down or up?  Yes, if an ML-based program receives all the data, which signals of the upcoming changes on the stock market. There are many factors needed to take into account, such as company’s current earnings reports, new products and acquisitions. External factors, such as news, economics, and politics trigger market to move as well. ML algos can process this enormous bunch of data and reveal hidden patterns, which can be of help to predict stock prices change.

Machine Learning also a brings value by looking to the past price movements. Such insights pulled from the past, could be really of help for analyzing current and predicting future market state.

Computer software may perform a task many times, and the result would always be the same. Without human intervening, the program’s output won’t change. But the application of Machine Learning is a more advanced solution, as it has great ability to self-learn over time. But do not think that if a program is able to educate itself, it could go out of control (other myths related to AI demystified here). It’s easy to set up the borders within which the software could not operate.

How will the future of hedge fund industry look like? According to Bloomberg, 58% of managers think machine learning will have a medium to large impact on the industry. Now, traders are still in the early stages of incorporating this powerful instrument. As for now, the opportunity is still untapped, whereas some companies have seized it and enjoyed the benefits. Undoubtedly, it is only a matter of time when the technology becomes widespread within the industry.

 

Are you interested in utilizing AI&ML for trading? Call us at +1 (973) 597-1000 or fill out the form below for a free consultation.

 

Your Name (required)

Your Email (required)