Trader Performance & Bitcoin Market Sentiment Analysis

Overview

This project analyzes the relationship between trader performance and Bitcoin market sentiment (Fear/Greed). It aims to uncover patterns that can guide trading strategies in Web3 markets.

Datasets Used

• Market Sentiment Data (market_sentiment.csv):

  • Columns: timestamp, value, classification, date
  • Source: Bitcoin Fear & Greed Index

• Trader Data (trader_data.csv):

  • Columns include: Account, Coin, Execution Price, Size USD, Side, Timestamp, Closed PnL, etc.
  • Source: Hyperliquid

Key Steps

1. Data Cleaning:

   • Parsed timestamps
   • Filtered out erroneous/unusable rows
   • Removed trades outside sentiment data range

2. Data Merging:

   • Joined trader data with sentiment classification on the date field

3. Exploratory Data Analysis (EDA):

   • Trade volume by sentiment classification
   • Average profit/loss under Fear vs Greed conditions

4. Insights:

   • Trader behavior varies significantly based on sentiment.
   • Higher PnL trends during certain sentiment conditions.

Tools & Libraries

• Python
• Pandas
• Matplotlib
• Seaborn
• Jupyter Notebook

Folder Structure

1. Clone the repository or download the project folder.
2. Set up a Python virtual environment and install dependencies:

   pip install -r requirements.txt