To make sure you never miss out on your favourite NEW stories, we're happy to send you some reminders

Click 'OK' then 'Allow' to enable notifications

Stock trader who returned 440% in 2022 explains how software he built works
Featured Image Credit: insta_photos / Agencja Fotograficzna Caro / Alamy Stock Photo

Stock trader who returned 440% in 2022 explains how software he built works

Gemy Zhou, the stock trader who founded the program, isn't sure if he'll be able to get the same results again.

A stock trader who returned 440 percent in 2022 has explained the system he built to automatically execute trades on his behalf.

Gemy Zhou entered the United States Investing Competition of 2022 during the worst year for stocks since 2008.

There, he finished in second place with a stock-trading computer program which helped him earn a massive 440.4 percent gain for the year.

Zhou moved to Canada after leaving China around 2004 and was living off of savings. He spent his time improving his data science and machine learning skills, using Coursera - an online education platform - for support.

Data science studies the relationships between different data points, and Zhou turned his attention to how he could use data science to trade. "I thought that maybe data science is a possible way to trade," Zhou told Business Insider. "So I just tested and made a lot of experiments and backtests. Although I didn't trade. but I did a lot of research."

Zhou turned his attention to how he could use data science to trade.
FALKENSTEINFOTO / Alamy Stock Photo

It would seem the odds were stacked against him initially. Zhou - who had owned a computer hardware business in China - previously tried trading Chinese stocks on the Shanghai Stock Exchange in 2004 with a process that was costly and time-consuming. He barely broke even, he said, after scouring through company data and earnings reports in a search for strong fundamentals.

This time around, Zhou used a general-purpose computer programming language called Python to gather stock data and scraped historical information spanning two decades from places like Yahoo Finance.

This gave Zhou around 20 to 30 data points, including information like moving-day averages for different lengths of time.

In a process called model training, the program can recognise which combinations can determine outcomes once the data points have been used to infer the relationship between their features and outcomes.

In this case of model training, the program determines the stock’s price. Now he used Interactive Broker’s data service to provide his program with stock information in real time.

PhotoEdit / Alamy Stock Photo

However, Norman Zadeh, the founder of the competition doesn’t think Zhou will be able to repeat his massive success.

Zadeh told Insider: "His performance is probably a combination of skill and luck.

“And so, more likely than not, he might return 30 percent next year. And I certainly would not suggest to people that you can just easily make 440 percent in the stock market. I think anybody who trades in the stock market should be happy if they make five or 10 percent."

Zadeh explained that finding which variables, whether interest rates or news, influence the market the most is something traders are always looking to discover and Zhou appears to have been more successful at attempting this than others.

He also concluded that Zhou essentially wrote a computer program that automatically generated trades.

At first, Zhou’s program was breaking even when he began testing the model in 2020 and 2021, however after a few adjustments things started to change massively.

New York Stock Exchange.
Image Professionals GmbH / Alamy Stock Photo

Post adjustments, Zhou’s program executes long and short positions. So, when the first order is placed to enter a position, the second order – or closing order – is placed at the same time at a calculated price.

Because positions can be held for an hour or once market closing approaches if the right conditions weren’t met, Zhou determines whether he wants to hold the stock overnight or manually sell it.

His program trades stocks priced under $1 a share to large caps and does about 20 to 50 trades a day, with the number of shares determined by a pre-set amount.

Zhou says he isn’t sure if he can get the 440 percent returns in the future as it depends on how the stock market changes and whether he can adjust his program to meet these changes.

A takeaway from the experience he shared is that as market conditions shift, so should your strategy.

Topics: Money