WaveQTE • Wavelet-based Quantile Transfer Entropy Analysis • WaveQTE

Wavelet-based Quantile Transfer Entropy Analysis for Financial Time Series

Advanced tools for analyzing market interconnectedness, financial contagion, and risk spillover effects across different time scales and market conditions.

🚀 Why WaveQTE?

WaveQTE bridges the gap between theoretical finance and practical analysis by combining wavelet decomposition with quantile transfer entropy to reveal hidden patterns in financial markets. Whether you’re a researcher, practitioner, or policymaker, WaveQTE provides the tools you need to understand:

📊 Market Interconnectedness - How markets influence each other across time scales
🌊 Financial Contagion - How crises spread between markets and regions
📈 Risk Spillovers - Directional risk transmission patterns
🔗 Network Dynamics - Complex relationships in financial systems

✨ Key Features

📁 Multi-source Data Support

Built-in financial dataset with 10 global market indices
Yahoo Finance integration for real-time data
Custom data import capabilities
Automated data quality checks

〰️ Wavelet Decomposition

MODWT-based analysis across multiple time scales
Robust handling of non-stationary financial data
Scale-specific analysis of market dynamics
Advanced variance decomposition

📊 Quantile Transfer Entropy

Information flow measurement across quantiles
Tail-risk and extreme event analysis
Directional spillover detection
Bootstrap significance testing

🕸️ Network Analysis

QTE-based network construction
Comprehensive network metrics
Community detection algorithms
Multi-scale network visualization

🌊 Financial Contagion

Crisis period identification
Cross-market correlation analysis
Dynamic contagion measurement
Regional spillover patterns

📈 Rich Visualizations

Interactive network plots
Publication-ready heatmaps
Time-varying analysis charts
Customizable themes and layouts

📦 Installation

# Install from GitHub (recommended)
if (!require("devtools")) install.packages("devtools")
devtools::install_github("avishekb9/WaveQTE")

# Load the package
library(WaveQTE)

Dependencies: WaveQTE automatically installs required packages including quantmod, waveslim, quantreg, igraph, ggplot2, and others.

🚀 Quick Start

1️⃣ Load Data

# Use built-in dataset (recommended for testing)
data <- get_stock_data(data_source = "builtin", 
                       indices = c("GSPC", "N225", "FTSE"))

# Or fetch live data from Yahoo Finance
# data <- get_stock_data(c("AAPL", "MSFT", "JPM"), 
#                        start_date = "2020-01-01", 
#                        end_date = "2023-12-31")

2️⃣ Process Returns

processed_data <- process_returns(data)
summary_stats <- calculate_summary_stats(processed_data)
print_summary_stats(summary_stats)

3️⃣ Wavelet Decomposition

# Decompose time series into multiple scales
wave_decomp <- lapply(1:ncol(processed_data), function(i) {
  wavelet_decompose(as.numeric(processed_data[1:256, i]), n.levels = 4)
})

# Analyze variance contributions across scales
var_analysis <- lapply(wave_decomp, analyze_wavelet_variance)
print_wavelet_summary(wave_decomp[[1]])

4️⃣ Quantile Transfer Entropy

# Calculate multiscale QTE
tau_levels <- c(0.1, 0.25, 0.5, 0.75, 0.9)
qte_results <- calculate_multiscale_qte(
  wave_decomp[[1]], 
  wave_decomp[[2]], 
  tau_levels
)

# Calculate multiscale QTE across quantiles
tau_levels <- c(0.1, 0.25, 0.5, 0.75, 0.9)
qte_results <- calculate_multiscale_qte(
  wave_decomp[[1]], wave_decomp[[2]], tau_levels
)

# Create beautiful heatmap visualization
plot_qte_heatmap(qte_results, title = "Multiscale QTE Analysis")

5️⃣ Network Analysis

# Build QTE-based financial network
network <- create_qte_network(processed_data[1:256, ], 
                             wave_decomp, scale = 1, tau = 0.5)

# Compute comprehensive network metrics
metrics <- calculate_network_metrics(network)
print(paste("Network density:", round(metrics$density, 3)))

# Create interactive network visualization
plot_enhanced_network(network, scale = 1, tau = 0.5)

6️⃣ Financial Contagion Analysis

# Analyze contagion between developed and emerging markets
dm_data <- get_stock_data(data_source = "builtin", 
                          indices = c("GSPC", "N225", "FTSE"))
em_data <- get_stock_data(data_source = "builtin", 
                          indices = c("SSE", "BSE", "HSI"))

# Measure contagion effects
contagion_results <- calculate_contagion_index(dm_data[1:256, ], 
                                              em_data[1:256, ])

# Identify crisis periods automatically
crisis_periods <- identify_crisis_periods(cbind(dm_data, em_data))

📚 Learning Resources

📖 Comprehensive Vignettes

Basic Usage - Get started with core functionality
Advanced Features - Explore sophisticated analysis techniques
Contagion Analysis - Specialized crisis and spillover analysis

📊 Data Sources

WaveQTE supports multiple data sources for maximum flexibility:

Source	Description	Use Case
Built-in	10 global market indices (2015-2023)	Testing and examples
Yahoo Finance	Real-time market data via quantmod	Live analysis
Custom CSV	Your own data files	Specialized datasets
Data Frames	Pre-loaded R objects	Custom workflows

# Examples of different data sources
builtin_data <- get_stock_data(data_source = "builtin")
yahoo_data <- get_stock_data(c("AAPL", "GOOGL"), "2020-01-01", "2023-12-31")
custom_data <- get_stock_data(data_source = "file", file_path = "my_data.csv")

🎯 Use Cases

🏛️ Academic Research

Publish-ready analysis for finance journals
Robust statistical methods with bootstrap testing
Comprehensive documentation and references

🏢 Risk Management

Real-time spillover monitoring
Crisis early warning systems
Portfolio diversification analysis

📊 Policy Analysis

Systemic risk assessment
Financial stability monitoring
Cross-border contagion analysis

💼 Investment Strategy

Market interconnectedness insights
Timing and allocation decisions
Risk-adjusted portfolio construction

🏗️ Package Architecture

WaveQTE/
├── 📁 R/                          # Core functionality
│   ├── 🔧 data_preparation.R       # Data import & processing
│   ├── 〰️ wavelet_decomposition.R # Multi-scale analysis
│   ├── 📊 quantile_transfer_entropy.R # QTE calculations
│   ├── 🕸️ network_analysis.R       # Network construction
│   ├── 🌊 contagion_analysis.R     # Crisis analysis
│   ├── 📈 visualization.R          # Rich visualizations
│   └── 🛠️ utils.R                 # Helper functions
├── 📚 vignettes/                   # Comprehensive tutorials
├── 🧪 tests/                       # Extensive test suite
└── 📖 man/                         # Complete documentation

🌟 Real-World Applications

Application	Description	Benefit
🏛️ Academic Research	Publish-ready analysis for finance journals	Rigorous methodology with statistical testing
🏢 Risk Management	Real-time spillover monitoring	Early warning systems for portfolio protection
📊 Policy Analysis	Systemic risk assessment	Evidence-based financial stability decisions
💼 Investment Strategy	Market interconnectedness insights	Enhanced portfolio diversification
🌍 Crisis Analysis	Cross-border contagion detection	Understanding global financial linkages

📈 Performance & Quality

✅ 109 passing tests with comprehensive coverage
✅ CRAN-compliant package structure
✅ Optimized algorithms for large datasets
✅ Bootstrap validation for statistical robustness
✅ Extensive documentation with real-world examples

🤝 Contributing

We welcome contributions! Whether you’re fixing bugs, adding features, or improving documentation:

🍴 Fork the repository
🌱 Create a feature branch
✨ Make your changes
🧪 Run tests (devtools::test())
📝 Submit a pull request

See CONTRIBUTING.md for detailed guidelines.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

📚 Citation

If you use WaveQTE in your research, please cite:

@misc{WaveQTE2024,
  author = {Avishek Bhandari},
  title = {WaveQTE: Wavelet-based Quantile Transfer Entropy Analysis for Financial Time Series},
  year = {2024},
  publisher = {GitHub},
  url = {https://github.com/avishekb9/WaveQTE},
  note = {R package version 0.1.0}
}

💬 Support & Community

🐛 Issues: GitHub Issues
📧 Contact: bavisek@gmail.com
📖 Documentation: Package Website

🙏 Acknowledgments

WaveQTE is built on the shoulders of giants:

R Core Team - The R statistical computing environment
waveslim package - Wavelet analysis functionality
igraph package - Network analysis capabilities
ggplot2 package - Beautiful and flexible visualizations
quantmod package - Financial data integration

⭐ Star this repository if WaveQTE helps your research! ⭐

Get Started • Documentation • Examples • GitHub