Stock Analytics API

A FastAPI-powered REST API for stock market analysis with machine learning features. Fetches real-time data from Yahoo Finance and provides price predictions, trend classification, and anomaly detection.

Features

• Real-time Stock Data — Fetch current stock info and historical prices via yfinance
• Price Prediction — Linear regression model to predict next-day closing price
• Trend Classification — Random Forest classifier to identify bullish/bearish trends
• Anomaly Detection — Z-score analysis to flag unusual price movements

Tech Stack

• Framework: FastAPI
• ML: scikit-learn (Linear Regression, Random Forest)
• Data: yfinance, pandas
• Server: Gunicorn + Uvicorn
• Deployment: Docker, Railway

Live Demo

Production URL: https://analytics-api-production-f8f1.up.railway.app

Try it:

• Health Check
• AAPL Stock Info
• AAPL 1 Month History
• AAPL Prediction
• AAPL Classification
• AAPL Anomalies
• API Docs

Endpoints

Health Check

GET /health

Response:

{
  "status": "ok"
}

---

Stock Info

GET /stocks/{symbol}

Example: GET /stocks/AAPL

Response:

{
  "symbol": "AAPL",
  "short_name": "Apple Inc.",
  "long_name": "Apple Inc.",
  "sector": "Technology",
  "industry": "Consumer Electronics",
  "current_price": 273.12,
  "market_cap": 4200000000000,
  "currency": "USD",
  "fifty_two_week_high": 280.50,
  "fifty_two_week_low": 164.08
}

---

Price History

GET /stocks/{symbol}/{period}

Example: GET /stocks/AAPL/1mo

Valid periods: 1d, 5d, 1mo, 3mo, 6mo, 1y, 2y, 5y, 10y, ytd, max

Response:

{
  "symbol": "AAPL",
  "period": "1mo",
  "prices": [
    {
      "date": "2024-12-01T00:00:00",
      "open": 270.00,
      "high": 275.50,
      "low": 269.25,
      "close": 273.12,
      "volume": 45000000
    }
  ]
}

---

Stock Stats

GET /stocks/{symbol}/{period}/stats

Example: GET /stocks/AAPL/1mo/stats

Response:

{
  "symbol": "AAPL",
  "period": "1mo",
  "avg_price": 268.45,
  "percentage_change": 5.2,
  "high": 280.50,
  "low": 255.00,
  "volatile": 8.32
}

---

Price Prediction

GET /stocks/{symbol}/{period}/predict

Example: GET /stocks/AAPL/1mo/predict

Predicts next-day closing price using Linear Regression.

Response:

{
  "symbol": "AAPL",
  "period": "1mo",
  "current_price": 273.12,
  "predicted_price": 276.24,
  "trend": "up"
}

---

Trend Classification

GET /stocks/{symbol}/{period}/classify

Example: GET /stocks/AAPL/6mo/classify

Classifies stock trend using Random Forest.

Response:

{
  "symbol": "AAPL",
  "period": "6mo",
  "trend": "bullish",
  "confidence": 68.0,
  "accuracy": 88.46,
  "probabilities": {
    "bearish": 32.0,
    "bullish": 68.0
  }
}

---

Anomaly Detection

GET /stocks/{symbol}/{period}/anomaly

Example: GET /stocks/AAPL/6mo/anomaly

Detects unusual price movements using z-score analysis.

Response:

{
  "symbol": "AAPL",
  "period": "6mo",
  "threshold": 2.0,
  "total_anomalies": 10,
  "anomalies": [
    {
      "date": "2024-11-15",
      "close": 225.50,
      "return_pct": -4.2,
      "z_score": -2.8,
      "type": "drop"
    }
  ],
  "note": "For educational purposes only"
}

Quick Start

Local Development

# Clone the repo
git clone https://github.com/KaanIsmet/Analytics-API.git
cd Analytics-API

# Create virtual environment
python -m venv venv
source venv/bin/activate

# Install dependencies
pip install -e .

# Run the server
uvicorn main:app --reload

Docker

docker build -t stock-analytics-api .
docker run -p 8000:8000 stock-analytics-api

Docker Compose

docker compose up

Disclaimer

This API is for educational purposes only. The predictions and analysis provided should not be used for actual trading decisions.

License

MIT