Methodological Insights into Structural Causal Modelling and Uncertainty-Aware Forecasting for Economic Indicators

2025-09-08

AI Summary

This study addresses the challenges of dynamic causal structure identification and uncertainty-aware forecasting for four key U.S. macroeconomic indicators: GDP, economic growth, inflation, and unemployment rate. Methodologically, it introduces a novel paradigm integrating causal discovery with zero-shot time-series forecasting: (i) latent causal structure is learned without prior assumptions using the LPCMCI framework augmented with Gaussian Process Distance Correlation (GPDC); (ii) the Chronos foundation model is leveraged for fine-tuning-free, uncertainty-quantified forecasting; and (iii) a novel confidence-interval deviation analysis is proposed to uncover latent influence mechanisms. Empirically, the work provides the first empirical validation of a unidirectional causal link from economic growth to GDP; achieves significant accuracy gains in 1–2-quarter-ahead unemployment forecasting; and demonstrates that 90% prediction intervals effectively support policy intervention and anomaly detection. The framework delivers an interpretable, robust, causal-predictive joint analysis tool for macroeconomic decision-making.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Game Theory and Economic Paradigms: Other Foundations of Game Theory & Economic Paradigms

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

This paper presents a methodological approach to financial time series analysis by combining causal discovery and uncertainty-aware forecasting. As a case study, we focus on four key U.S. macroeconomic indicators -- GDP, economic growth, inflation, and unemployment -- and we apply the LPCMCI framework with Gaussian Process Distance Correlation (GPDC) to uncover dynamic causal relationships in quarterly data from 1970 to 2021. Our results reveal a robust unidirectional causal link from economic growth to GDP and highlight the limited connectivity of inflation, suggesting the influence of latent factors. Unemployment exhibits strong autoregressive dependence, motivating its use as a case study for probabilistic forecasting. Leveraging the Chronos framework, a large language model trained for time series, we perform zero-shot predictions on unemployment. This approach delivers accurate forecasts one and two quarters ahead, without requiring task-specific training. Crucially, the model's uncertainty-aware predictions yield 90% confidence intervals, enabling effective anomaly detection through statistically principled deviation analysis. This study demonstrates the value of combining causal structure learning with probabilistic language models to inform economic policy and enhance forecasting robustness.

Problem

• Uncovering dynamic causal relationships among macroeconomic indicators
• Performing zero-shot probabilistic forecasting for unemployment
• Enabling anomaly detection through uncertainty-aware confidence intervals

Innovation

• Combining causal discovery with uncertainty-aware forecasting
• Applying LPCMCI framework with GPDC for causal relationships
• Using Chronos LLM for zero-shot probabilistic forecasting