Can a set of math equations predict where a stock price is headed? You gather earnings data from SEC filings for over 108 stocks across different sizes and industries. Each financial indicator gets a score from -1 to +1.
You factor these scores into an overall equation. Then you test the system's predictions over time periods from 3 months to 5 years.
The resulting system identifies profitable stocks about 95% of the time. Over a one-year test period, the picks return 64% on the investment with less than 5% prediction error.
Hypothesis
The hypothesis is that a system using theoretical and mathematical ratios and data can predict future profitable stocks 95% of the time.
Messy real-world data becomes a testable prediction when you turn it into a scoring system. You gather earnings data from SEC filings for over 108 stocks across different sizes and industries, then assign each financial indicator a score from -1 to +1. Those scores feed into an overall equation that copies the behavior of real stock prices. Testing that equation over time periods from 3 months to 5 years shows whether the model's predictions hold up against actual market results.
Can a set of step-by-step rules predict where a stock price is headed? You gather earnings data from SEC filings for over 108 stocks across different sizes and industries. Each financial indicator gets a score from -1 to +1, and those scores feed into an overall equation — the same fixed rules applied the same way each time. When you test the system's predictions over periods from 3 months to 5 years, it identifies profitable stocks about 95% of the time. Over a one-year test period, the picks return 64% on the investment with less than 5% prediction error.
Method & Materials
You will use data from SEC filings of stocks of different sizes and industries to gather fundamental data regarding their earnings. You will then factor a value of -1 to +1 for each indicator into the overall equation.
You will need data from SEC filings of stocks of different sizes and industries, a calculator, and a computer.
MEL Math — hands-on math experiment kits delivered monthly — makes abstract concepts tangible. (Affiliate link)
This project has shown that a system using theoretical and mathematical ratios and data can predict future profitable stocks 95% of the time. The result of this project was a system which can predict the future fluctuations and trends of a security with less than a 5% error.
Why do this project?
This science project is unique because it provides an opportunity to use math and data to predict stock market fluctuations and make money.
Also Consider
Experiment variations to consider include predicting price fluctuations of securities over the course of 3 months, 6 months, 1 year, 3 years and 5 years.
Full project details
Additional information and source material for this project are available below.
