Skip to content

README.md

Latest commit

 

History

History
98 lines (72 loc) · 3.16 KB

README.md

File metadata and controls

98 lines (72 loc) · 3.16 KB

Python Stock Prediction Tool

A desktop stock prediction and tracking app built with PyQt6 and SQLite3. The app provides technical analysis-based predictions using Bollinger Bands, RSI, Moving Averages, Volume indicators, and live data & news. It helps users make informed decisions on buying, selling, or holding stocks.

Features

  • PyQt6 GUI for interactive desktop experience
  • SQLite3 database to store stock data
  • Fetches live stock data using yfinance
  • Fetches live business news from NewsData API
  • Calculates key technical indicators:
    • RSI (Relative Strength Index)
    • Bollinger Bands
    • Moving Averages (SMA 20, 50, 200)
    • Volume analysis
  • Generates buy/sell/hold recommendations with confidence scores
  • Computes target prices based on volatility and moving averages

Stock Prediction Logic

1. Data Collection

  • Historical stock data (1 year of daily closing prices)
  • Trading volume
  • Simple Moving Averages (20-day, 50-day, 200-day)
  • Live stock data using yfinance

2. Technical Indicators

Indicator Description Usage
RSI Momentum oscillator (0–100) RSI < 30 → bullish, RSI > 70 → bearish
Bollinger Bands Volatility bands around SMA 20 Price near upper band → overbought, lower band → oversold
Moving Averages SMA 20, 50, 200 Golden Cross (SMA 50 > SMA 200) → bullish, Death Cross → bearish
Volume Trading volume analysis Unusually high volume → potentially bullish

3. Score Calculation System

Component Condition Score
Bollinger Bands Price ≤ lower band +0.3
Price ≥ upper band -0.3
RSI <30 +0.25
>70 -0.25
Moving Average Cross Golden Cross +0.25
Death Cross -0.25
Volume High volume (>1.5× avg) +0.2
  • Total score ranges -0.8 (extremely bearish) to +0.8 (extremely bullish)

4. Prediction Categories

Score Recommendation
≥ 0.5 Strong Buy
0.15 – 0.5 Buy
-0.15 – 0.15 Hold
-0.5 – -0.15 Sell
≤ -0.5 Strong Sell

5. Target Price Calculation

  • Uses 20-day SMA as base and price standard deviation for volatility.
  • Strong Buy: SMA 20 + 70% of SD
  • Buy: SMA 20 + 30% of SD
  • Hold: Current price
  • Sell: SMA 20 - 30% of SD
  • Strong Sell: SMA 20 - 70% of SD

Two types of standard deviation used:

  • Rolling SD (Bollinger Bands): short-term, adapts to recent trends
  • Global SD (Target Price): long-term, stable measure of overall volatility

Live News Feature

  • Fetches business news from NewsData API
  • Categorizes news into Market Updates, Investing, Company News, Regulations, Economy, Crypto, Banking, Others
  • Allows searching news by keywords
  • Displayed in a scrollable PyQt6 GUI

Installation

# Clone the repo
git clone git@github.com:runt1meerr0r07/Python_Stock_Prediction_Tool.git
cd Python_Stock_Prediction_Tool

# Install dependencies
pip install PyQt6 pandas numpy matplotlib yfinance requests