GARJI volatility-based predictive causal networks in corporate credit markets
Paper introduces a GARJI-Granger volatility model and QUBO optimization for constructing predictive causal networks in corporate credit markets.
What it examines
The paper introduces a predictive causality network among corporate bond issuers using Granger causality tests on volatility estimates from a GARJI model. It constructs a Directed Acyclic Graph and applies QUBO optimization to enforce acyclicity and clustering. This tool aids proactive portfolio management and risk assessment in credit markets.
What it concludes
The study finds that using GARJI-based volatility measures with Granger causality and QUBO optimization enhances predictive networks for corporate credit markets. This approach improves portfolio management, risk monitoring, and early warning of contagion. Potential applications include asset management, systemic risk analysis, and diversification strategies, with future research exploring non-linear extensions.
Evidence objects
Researchers build a groundbreaking causal network among corporate bond issuers using $$\text{GARJI}$$-derived volatility estimates, effectively filtering noise and capturing jump dynamics to enhance Granger causality tests and market risk prediction.
key_findings bullet 1 · key_findings · validation V0
Employing QUBO optimization, researchers construct a Directed Acyclic Graph that predicts risk contagion and outperforms conventional models through improved accuracy, successfully signaling market stress during COVID-19 and the Russo-Ukrainian conflict.
key_findings bullet 2 · key_findings · validation V0
The study redefines volatility estimation by integrating jump components and a father-son relationship concept, with backtesting demonstrating robust out-of-sample performance despite parameter sensitivity and market friction omissions, promising practical applications.
key_findings bullet 3 · key_findings · validation V0
The paper integrates $GARJI$ volatility modeling, Granger-causality networks, and the $QUBO$ optimization framework into a novel tool for managing corporate bond portfolios and monitoring systemic risk. It bridges established methods with innovative applications in corporate credit, providing fresh insights and practical solutions for fixed income market challenges, undeniably impactful contribution.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
- … a pletora of stylized facts about financial series. This network … cally significant and computationally efficient, which … finance literature, connecting theory of financial networks …
Source row: 959 · abstract type: snippet