Skip to content

Darkest-Teddy/StalkMarket

Repository files navigation

Stalk Market: Learn Investing Through Growth

About Our Project

Financial markets are complex and intimidating, making it hard for beginners to learn how investing works.
Stalk Market simplifies financial education by turning investing into an interactive and visual experience — where stocks grow like plants.
Our platform helps users learn key financial principles safely, through simulation, gamification, and guided experimentation.

At the core of our system is a hybrid predictive model combining traditional algorithmic forecasting with AI-based prediction:

  • Algorithm-Based Model: Uses macroeconomic data (CPI, unemployment rate, etc.) and the Geometric Brownian Motion (GBM) model from the FRED dataset to estimate stock movements.
  • AI-Based Model: Trained with 10 years of historical stock data from Yahoo Finance, using transfer learning on a modified GPT-2 model.
    • GPT-2 is adapted to handle numerical sequences: previous price fluctuations as input, and predicted price changes as output.
  • Final Price: Computed as
    0.8 × (Algorithm Prediction) + 0.2 × (AI Prediction)

To make learning engaging and realistic, we also introduce:

  • Random Market Events: These simulate real-world unpredictability — affecting both crop (stock) growth and prices.
  • Education Mode: An extra ML layer discourages risky investing by adjusting market volatility. If a user “puts all eggs in one basket,” the model reduces volatility to show why diversification matters.

Our platform is designed for accessibility and fun. Users track their portfolio as a garden — with plants that grow, bear fruit, or wither — visually showing how decisions and market conditions impact performance. Fonts are large, navigation is simple, and visuals are engaging enough for all ages.

Running the Project

Option 1: Local development

cd path/to/StalkMarket
python -m uvicorn main:app --reload --port 8000

Frontend assets live in public/; open public/index.html or run a static server (e.g., python -m http.server 5173).

Accomplishments

  • Developed a working hybrid AI + algorithmic stock simulation
  • Built a stable and accessible web platform suitable for all learners
  • Education mode to reinforce responsible investing

What's Next

  • Add personalized learning paths that adapt to user choices and skill levels
  • Introduce social and collaborative features — e.g., classroom mode or friendly competitions
  • Expand random event types (economic news, weather, policy shifts)

About

Simulate the stock market in a more colorful way

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors