Temporal-Aligned Meta-Learning for Risk Management: A Stacking Approach for Multi-Source Credit Scoring

TL;DR

This paper introduces a novel meta-learning framework that aligns temporal data in credit risk assessment for SMEs, improving prediction stability and interpretability by integrating multiple scoring systems and addressing data delays.

Contribution

It proposes a two-step temporal decomposition and stacking architecture for credit scoring, effectively handling temporal misalignment and combining diverse models without retraining.

Findings

Enhanced temporal consistency in credit risk predictions

Improved predictive stability over standard ensemble methods

Effective modeling of credit risk evolution over time

Abstract

This paper presents a meta-learning framework for credit risk assessment of Italian Small and Medium Enterprises (SMEs) that explicitly addresses the temporal misalignment of credit scoring models. The approach aligns financial statement reference dates with evaluation dates, mitigating bias arising from publication delays and asynchronous data sources. It is based on a two-step temporal decomposition that at first estimates annual probabilities of default (PDs) anchored to balance-sheet reference dates (December 31st) through a static model. Then it models the monthly evolution of PDs using higher-frequency behavioral data. Finally, we employ stacking-based architecture to aggregate multiple scoring systems, each capturing complementary aspects of default risk, into a unified predictive model. In this way, first level model outputs are treated as learned representations that encode non-linear relationships in financial and behavioral indicators, allowing integration of new expert-based features without retraining base models. This design provides a coherent and interpretable solution to challenges typical of low-default environments, including heterogeneous default definitions and reporting delays. Empirical validation shows that the framework effectively captures credit risk evolution over time, improving temporal consistency and predictive stability relative to standard ensemble methods.