Paper: arXiv 2408.15404

Abstract

This paper presents the experimental process and results of SVM, Gradient Boosting, and an Attention-GRU Hybrid model in predicting the Implied Volatility of rolled-over five-year spread contracts of credit default swaps (CDS) on European corporate debt during the quarter following mid-May ‘24, as represented by the iTraxx/Cboe Europe Main 1-Month Volatility Index (BP Volatility). The analysis employs a feature matrix inspired by Merton’s determinants of default probability. Our comparative assessment aims to identify strengths in SOTA and classical machine learning methods for financial risk prediction

Complexity vs Empirical Score

  • Math Complexity: 7.2/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced architectures like Attention-GRU hybrids and includes mathematical formalisms (e.g., volatility index formula, Merton model references), indicating moderate-to-high complexity, while its experimental design with comparative ML models and feature engineering demonstrates substantial empirical backing.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Predict CDS Implied Volatility<br>using SOTA ML methods"] --> B["Data Preparation<br>Merton Determinants Feature Matrix<br>iTraxx Europe Main 1-Month Volatility Index"]
  B --> C{"Model Selection"}
  C --> D["Support Vector Machine SVM"]
  C --> E["Gradient Boosting"]
  C --> F["Attention-GRU Hybrid"]
  D & E & F --> G["Experimental Process:<br>Incremental Batch Learning<br>Validation"]
  G --> H["Key Findings"]
  H --> I["Strengths identified for<br>Financial Risk Prediction"]
  H --> J["Performance Comparison:<br>SOTA vs Classical ML"]