We like games partly because of the uncertainty and surprises that they offer. The stock market is very much like a game in that context. AI offers a vast computational power to use real-time prices available of various stocks and built neural networks to take advantage of human sentiment for stock price prediction. However, many algorithms have not been very successful who deployed in real-life scenarios.
Eric Engle has discussed this in his research paper titled “The stock market as a game: An agent based approach to trading in stocks” which forms the basis of the following text.
Importance of this Research for Stock Market Trading
The researcher has proposed an ML model that analyses stock price based on the Fundamental Analysis of the company that the stock represents. The underlying hypothesis is that the agents trading based on fundamental analysis will do better than those using technical analysis over the long term.
Context of the Research Paper
In the words of the researcher,
If ever there were a field in which machine intelligence seemed destined to replace human brainpower, the stock market would have to be it. Investing is the ultimate numbers game, after all, and when it comes to crunching numbers, silicon beats gray matter every time. Nevertheless, the world has yet to see anything like a Wall Street version of Deep Blue, the artificially intelligent machine that defeated chess grand master Gary Kasparov in 1997. Far from it, in fact: When artificial-intelligence-enhanced investment funds made their debut a decade or so ago, they generated plenty of media fanfare but only uneven results. Today those early adopters of AI, like Fidelity Investments and Batterymarch Financial, refuse to even talk about the technology…Data flows in not just from standard databases but from everywhere: CNN, hallway conversations, trips to the drugstore. ‚Unless you can put an emotional value on certain events and actions, you can’t get the job done.‘ Naturally, investors don’t process this hodgepodge of inputs according to some set of explicit, easily transcribed rules. Instead, the mind matches the jumble against other jumbles stored in memory and looks for patterns, usually quite unconsciously. ‚Often, great investors can’t articulate the nature of their talent. They’re like pool players who make incredible trick shots on intuition.‘ Fine for them, but how do you code that?
Types of Stock-trading Agents
The researcher has described 6 agents for Stock Trading. Kindly note that the below 6 agents do not learn, and they are as follows:
Bears: Agents defined as Bears looks at several factors such as favorable PE ratio (Price to earnings ratio) < 30, low debt-equity ratio, the company value as per stock price < company value as per balance sheet, positives earnings, etc. to make purchases
Conservatives: This agent would buy if the PE ratio is favorable (<30) and the company is undervalued, that is, book value < market value
Blue Chip Investor: If the company’s dividend is more than 1$ per share, the agent will buy. If it falls below 1, the agent will sell.
Bargain Hunters: If the company value as determined by the stock market is less than its book value, this agent would qualify it as an undervalued stock and buy. Similarly, vice versa for sale.
Fools: This agent will make buy/sell decisions based on the price-to-earning ratio of the share. If the Price to Earnings (PE Ratio) is < 30, the agent will buy the share. When PE goes above 30, it will sell.
Idiots: This agent will make buy & sell based on the overall market sentiment & will assume that the market sentiment, in general, would not change in a day. So, if the market goes up today, this agent would buy, presuming that tomorrow would also be the same & vice-versa.
Research Result
The researcher has suggested a combined approach based on the above agents and referred to it as Eric Agent. The Eric Agent only buys when:
Stocks with PE < 30
Stocks that are lower than or within 10% of their book value
Stocks from companies that are making profits this year
Only buy stocks with < 1 debt to equity ratio
For Selling, Eric agent would sell any shares that have gained 20% from their purchase price.
Conclusion
The researchers accept that the proposed Eric agent will miss some profit-making and trading opportunities, but this conservative approach would also help the agent avoid losses. The researcher has mentioned that most multiplier games are zero-sum games, meaning if 1 player wins, the other will lose by default. However, when played over the long term, stock markets could be a positive-sum game, and the researcher has proposed Eric Agent do the same.
Future Work
If the proposed Eric agent by the researchers also starts using insights from ML models built on technical analysis, it could be further improved. If other factors such as Inflation, Oil Prices, and decline in productivity are also included, it would further enhance the Eric agent’s accuracy.
Source: Eric Engle’s “The stock market as a game: An agent based approach to trading in stocks“
You must be logged in to post a comment.