FinCARE: Financial Causal Analysis with Reasoning and Evidence

2025-10-23

AI Summary

Traditional portfolio analysis relies on correlation, limiting its ability to uncover causal mechanisms driving performance. This paper proposes a hybrid framework integrating statistical causal discovery with domain-specific financial knowledge to enable interpretable and verifiable counterfactual reasoning and intervention analysis. Methodologically, we construct a financial knowledge graph from SEC 10-K filings and synergistically leverage large language models (LLMs) for causal hypothesis generation and logical validation; we further enhance robustness by ensembling three causal discovery algorithms—PC, GES, and NOTEARS. Evaluated on synthetic data from 500 firms, our approach improves F1 scores by 36%–366% across algorithms, achieves a mean absolute error of 0.003610 in counterfactual prediction, and attains 100% accuracy in intervention direction identification. The core contribution is a novel knowledge graph–LLM co-driven causal discovery paradigm that jointly ensures statistical validity and domain plausibility.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Portfolio managers rely on correlation-based analysis and heuristic methods that fail to capture true causal relationships driving performance. We present a hybrid framework that integrates statistical causal discovery algorithms with domain knowledge from two complementary sources: a financial knowledge graph extracted from SEC 10-K filings and large language model reasoning. Our approach systematically enhances three representative causal discovery paradigms, constraint-based (PC), score-based (GES), and continuous optimization (NOTEARS), by encoding knowledge graph constraints algorithmically and leveraging LLM conceptual reasoning for hypothesis generation. Evaluated on a synthetic financial dataset of 500 firms across 18 variables, our KG+LLM-enhanced methods demonstrate consistent improvements across all three algorithms: PC (F1: 0.622 vs. 0.459 baseline, +36%), GES (F1: 0.735 vs. 0.367, +100%), and NOTEARS (F1: 0.759 vs. 0.163, +366%). The framework enables reliable scenario analysis with mean absolute error of 0.003610 for counterfactual predictions and perfect directional accuracy for intervention effects. It also addresses critical limitations of existing methods by grounding statistical discoveries in financial domain expertise while maintaining empirical validation, providing portfolio managers with the causal foundation necessary for proactive risk management and strategic decision-making in dynamic market environments.

Problem

• Portfolio managers lack true causal relationship analysis methods
• Existing approaches fail to capture performance-driving causal factors
• Current methods rely on correlation-based and heuristic analyses

Innovation

• Integrates causal discovery with financial knowledge graphs
• Enhances algorithms using LLM reasoning for hypotheses
• Combines statistical methods with domain expertise constraints