Ticker Data Analysis and Visualization

This repository contains a comprehensive data analysis pipeline for analyzing and visualizing stock ticker data. The primary focus is on NVIDIA (NVDA) stock, with features to adjust for stock splits, normalize data, and compute various statistics for financial analysis.

---

Features

1. Data Loading

• Utilizes GPU acceleration for efficient data handling using cuDF.
• Processes multiple CSV files containing historical trading data.

2. Stock Split Adjustment

• Automatically adjusts historical price data for stock splits based on NVIDIA's split history.

3. Data Transformation

• Computes daily returns.
• Logarithmically transforms and scales data for better analysis.

4. Statistical Analysis

• Calculates key statistics like:
  • Mean, Median, Std Dev, Skewness, Kurtosis
  • Value at Risk (VaR)
  • Sharpe Ratio
  • Jarque-Bera test for normality.

5. Data Visualization

• Creates insightful plots:
  • Adjusted and unadjusted price time series.
  • Price distributions.
  • Daily returns distributions.
  • Hourly price volatility.
  • Log-transformed counterparts of the above.

6. Data Normalization

• Normalizes prices to a range of 0-1 with configurable buffer.
• Supports scaling log-transformed prices.

---

Dependencies

• Python Libraries:

  • yfinance
  • pandas
  • matplotlib
  • cuDF (GPU acceleration)
  • torch
  • seaborn
  • scikit-learn
  • numpy
  • scipy

• Google Colab (optional):

  • Includes integration for Google Drive for seamless file handling.

---

File Breakdown

• Imports and Setup: Handles dependencies and GPU checks.
• Data Directory Config: Defines file paths for training and testing data.
• Stock Split Adjustment: Adjusts price data for historical splits.
• Visualization: Generates multi-panel plots to interpret data trends and anomalies.
• Statistical Analysis: Computes metrics to understand return distributions and risks.

---

Usage

1. Set Up the Environment

• Clone the repository and ensure all dependencies are installed.
• Optionally mount Google Drive for CSV file access.

2. Customize Configuration

• Modify settings in the Config class to analyze different tickers or date ranges.

3. Run the Notebook

• Execute the provided script to analyze and visualize stock ticker data.

4. Analyze Results

• Review output plots and statistical summaries for insights into the stock's historical performance.

---

Sample Outputs

Plots

• Split-adjusted price time series with split markers
• Price distribution histograms
• Daily returns distribution
• Volatility over time

Statistical Summaries

• Basic stats: Mean, Median, Std Dev, etc.
• Risk metrics: VaR, Sharpe Ratio.
• Normality test: Jarque-Bera results.

---

Future Enhancements

• Add support for multiple tickers.
• Extend analysis to include more financial metrics (e.g., Moving Averages, RSI).
• Improve visualizations with interactive tools (e.g., Plotly).

---

License

This project is open-source and available under the MIT License.

MYM-Aa

QAlgo Trade