Go to NeurIPS 2025 Workshop GenAI in Finance homepageFinCARE: Financial Causal Analysis with Reasoning & Evidence
Keywords: causal discovery, causal inference, financial modeling, knowledge graphs, large language models, counterfactual analysis, portfolio management, risk management, agentic AI
TL;DR: We combine statistical causal discovery with financial knowledge graphs and LLMs to build better causal models for portfolio management, achieving 43% better counterfactual predictions than traditional correlation-based methods.
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 optimiza-
tion (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 algo-
rithms: 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 counterfac-
tual predictions and perfect directional accuracy for intervention effects. It also
addresses critical limitations of existing methods by grounding statistical discov-
eries in financial domain expertise while maintaining empirical validation, pro-
viding portfolio managers with the causal foundation necessary for proactive risk
management and strategic decision-making in dynamic market environments.
Submission Number: 99
Loading