AI-Assisted Analysis of Policy Communication during Economic Crises: Correlations with Market Confidence and Recovery Outcomes

Aixin Kang, Kai Zhang and Yuexing Chen

Pinnacle Academic Press Proceedings Series, 2025, vol. 3, 159-173

Abstract: Economic crises necessitate effective policy communications to stabilize markets and facilitate recovery. This study introduces a novel multi-modal natural language processing (NLP) framework for analyzing policy communications during five major economic crises (2008-2009 Global Financial Crisis, 2010-2012 European Debt Crisis, 2015-2016 Commodity Crash, 2020 COVID-19 Downturn, and 2022-2023 Inflation Surge). The framework integrates the Flying Neural Network architecture - a novel neural model designed for dynamic pattern recognition - with reinforcement learning mechanisms to identify linguistic features influencing market confidence. Analysis of 14,297 official policy statements across seven major economies reveals significant correlations between communication characteristics and market confidence indicators. Commitment signaling emerges as the most influential linguistic feature (importance score 0.37), with maximum correlation strength occurring two days after the announcement (T + 2). Communications exhibiting high clarity indices (CI > 0.65) and moderate technical density (TD 0.30-0.45) demonstrated superior effectiveness in stabilizing market expectations. Temporal analysis indicates systematic variation in optimal communication strategies across crisis phases, with balanced approach communications generating most favorable recovery trajectories (average duration: 34.3 weeks). The reinforcement learning model achieves 83.7% directional forecast accuracy and 71.4% accuracy in recovery duration estimates. These findings advance the theoretical understanding of economic communication dynamics while providing actionable guidelines for optimizing policy communications during crisis periods.

Keywords: policy communication analysis; economic crisis management; market confidence modeling; Flying Neural Network (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://pinnaclepubs.com/index.php/PAPPS/article/view/190/192 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:dba:pappsa:v:3:y:2025:i::p:159-173

Access Statistics for this article

More articles in Pinnacle Academic Press Proceedings Series from Pinnacle Academic Press
Bibliographic data for series maintained by Joseph Clark ().