This stock trader explained how he managed to end 2022 with an incredible 440% appreciation

Gemy Zhou is 99% unknown to you as an investor, I totally understand that, he never really excelled as an investor and he was never very good at picking individual titles. But where did the giant change occur that made him one of the top investors of 2022, where he was able to gain 440% appreciation in a challenging market environment?

Gemy Zhou

Gemy Zhou came in second place in the stock division for the 2022 US Investment Contest with an incredible 440% appreciation! He was among 326 participants from around the world who tested their trading skills during the worst year for stocks since 2008.

But Zhou had a big advantage over the others: his knowledge of data science and coding. He entered the competition to test a computer program for stock trading that he had spent several years developing and debugging. The program then spent the year executing trades on his behalf automatically, earning him a whopping 440.4% profit for the year, according to his monthly brokerage statements.

  • Many, however, say that this is a unique feat that just can't be repeated.

"His performance is probably a combination of skill and luck," experts say.

The creator of the competition (Zadeh) for the best investor said:

"Certainly no one should think that you can easily make 440% in the stock market. I think anyone who trades in the stock market should be lucky if they make five or 10%."

Zadeh added that anyone who can make significant returns in the market should be particularly proud of themselves because the stock market is not a completely level playing field.

How did the programme come about in the first place?

Gemy Zhou was always interested in anything that had the potential to make money. And since data science studies the relationships between different data points, the stock market seemed like a good place to test his skills. After all, traders use a variety of indicators to try to predict the direction of prices, he said. By 2020, he was experimenting with data and coding programs that could trade successfully.

"I thought maybe data science is a possible way to trade," Zhou said. "So I just tested and did a lot of experiments and back-testing. Even though I didn't trade, I did a lot of research."

Building

Zhou said he initially collected stock data using Python, a general-purpose computer programming language, to get 20 years' worth of historical information from sites like Yahoo Finance. The result was about 20 to 30 data points, including things like moving day averages for varying lengths of time. The data is used to infer the relationship between their characteristics and the results. Zhou noted that this process is called model training and allows the program to recognize which combinations can determine outcomes, in this case the stock price. Today, he uses Interactive Brokers' data service to feed his program with real-time stock information that doesn't lag.

But even after creating the program, Zhou says he still doesn't know which variables have more weight or impact on the machine's decision to execute a trade. Because he was a trader in the past, he could make educated assumptions about which variables might be less or more important given market conditions, he said.

When Zhou first started testing the model in 2020 and 2021, it was groundbreaking until he made a few adjustments, he said. The main one was to reduce the weight that short-term price movements had on the program's decision to make a trade. He did this by raising the threshold of the amount of time a price has to move before the program reacts to avoid triggering an early trade. This adjustment was particularly important in the highly volatile market of 2021. But when the market slowed last year, he lowered the threshold a bit.

The program executes long and short positions. When the first order is entered to enter a position, the second order (the close order) is also entered at the same time at the calculated price. Positions can be held for one hour or until the market closes.

The program trades stocks ranging from penny stocks priced under $1 per share to large capitalization stocks and executes about 20 to 50 trades per day.

Here's how the program works

  • Gemy Zhou created a program that trades based on 20 to 30 data variables.
  • He fed the computer 20 years of data to determine the relationship between certain data points.
  • Its job is to maintain top performance by adapting to market conditions and calculating its own optimal prices - buying x selling.

Examples of trades

For example, on May 11, the program received a price change for Armstrong Flooring Inc, stock symbol AFI, which has since been delisted. The program evaluated the value, which triggered a buy signal for 7,500 shares. The system immediately entered a buy order at the market ask price of $0.3004 and also simultaneously entered an opposite closing sell order at a price of $0.3449, which was calculated based on many factors such as market volatility and historical volatility of the stock. Approximately 20 minutes later, the market price rose and reached the sell order price of $0.3449 and the position was closed.

On February 17, 2022 at 9:31 a.m., Knowbe4 Inc (KNBE) experienced a significant price change. The system immediately entered a sell order on 49 shares at a market bid price of $24.57. At the same time, it entered a short order with a closing price target of $21.79, which was again calculated based on many factors such as market volatility and historical volatility of the stock. About 17 minutes later, the market price dropped and hit the buy order price and the position was closed.

Conclusion

Zhou essentially wrote a computer program that automatically generates trades. This is similar to what quantitative hedge funds like Renaissance Capital do on a much larger scale.

Zhou says the key takeaway from his experience is that regardless of whether you use a program or trade manually, you have to backtest your theory. The second thing he has learned is that market conditions are extremely important. As those conditions change, your strategy should too, so you have to keep upgrading the program to make the data as accurate as possible for evaluating trades.

  • Friends I don't know about you, but I want that program! Feel free to let me know in the comments if you would use such a program 😄

Read the full article for free?
Go ahead 👇

Do you have an account? Then log in . Or create a new one .

No comments yet
Timeline Tracker Overview