Forecasting Climate Policy Uncertainty: Evidence from the United States

2025-07-16

AI Summary

This paper addresses the challenge of forecasting U.S. climate policy uncertainty (CPU), whose unpredictability impedes green investment, complicates regulatory planning, and intensifies public opposition. We propose a high-dimensional Bayesian structural time series (BSTS) model integrating macroeconomic indicators, financial cycle data, and Google Trends–derived public sentiment. A novel dynamic feature selection mechanism—based on spike-and-slab priors—automatically identifies salient drivers of CPU, while impulse response analysis quantifies the time-varying effects of macrofinancial shocks. Empirical evaluation demonstrates that our model significantly outperforms conventional econometric and deep learning benchmarks in medium- to long-horizon multi-step forecasting, achieving superior stability and accuracy. By delivering interpretable, robust, and forward-looking insights, the framework supports evidence-based climate governance and green investment decision-making.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Planning, Routing, and Scheduling: Scheduling under Uncertainty

Application Category

• Economics, Online Markets and Human Computation: Sustainability of Web economics
• Web Mining and Content Analysis: Bridging structured and unstructured data
• Responsible Web: Sustainability and climate impact of web technologies

Abstract

Forecasting Climate Policy Uncertainty (CPU) is essential as policymakers strive to balance economic growth with environmental goals. High levels of CPU can slow down investments in green technologies, make regulatory planning more difficult, and increase public resistance to climate reforms, especially during times of economic stress. This study addresses the challenge of forecasting the US CPU index by building the Bayesian Structural Time Series (BSTS) model with a large set of covariates, including economic indicators, financial cycle data, and public sentiments captured through Google Trends. The key strength of the BSTS model lies in its ability to efficiently manage a large number of covariates through its dynamic feature selection mechanism based on the spike-and-slab prior. To validate the effectiveness of the selected features of the BSTS model, an impulse response analysis is performed. The results show that macro-financial shocks impact CPU in different ways over time. Numerical experiments are performed to evaluate the performance of the BSTS model with exogenous variables on the US CPU dataset over different forecasting horizons. The empirical results confirm that BSTS consistently outperforms classical and deep learning frameworks, particularly for semi-long-term and long-term forecasts.

Problem

• Forecasting Climate Policy Uncertainty in the US
• Managing economic and environmental policy trade-offs
• Improving CPU predictions using Bayesian time-series models

Innovation

• Bayesian Structural Time Series model
• Dynamic feature selection mechanism
• Impulse response analysis validation