Shockprint Mapping and Systemic Resonance: A Novel Fuzzy Framework for Cross-Asset Volatility and Crisis Prediction in Traditional, ESG, and Crypto Markets (2017–2025)

The global investment behavior has been mainly influenced by macroeconomic forces such as the dominating impact of the COVID-19 pandemic, post-COVID-19 geopolitical tensions, and deepening climate-related risks. These factors have a multitude of impacts in terms of shocks on the asset investments, often disregarded by traditional asset allocation models. The present study proposes a new machine learning framework for prediction of shifts in preferences of investors for major asset classes that includes equities, bonds, gold, cryptocurrencies, and ESG-linked instruments under the shadow of uncertainties in the global market. The proposed model integrates macroeconomic indicators, geopolitical risk indices, and financial data sentiments to identify pre-impact signals of asset rotation and ascertain the decomposed financial resilience at the earliest. Forecasting techniques such as Random Forest and XGBoost have been combined with SHAP for interpretability and sentimental analysis for determining positive, negative, or neutral sentiments for asset class investments. The empirical results derived from a time series dataset traversing from 2017 to 2025 uncover meaningful investor buying behavior patterns during major shocks, presenting probable insights benefiting institutional investors, portfolio strategists, and policymakers focused on market stability and sustainable finance. The research emphasizes the need for flexible and accommodating AI-driven strategies for tackling complexities in highly complex poly-dimensional markets using the structure equation model based on empirical analysis.