ACT-Tensor: Tensor Completion Framework for Financial Dataset Imputation

2025-08-03

AI Summary

Financial panel data frequently exhibit severe and heterogeneous missingness, undermining the performance of conventional imputation methods—particularly in modeling multidimensional structures, capturing cross-sectional heterogeneity, and handling extremely sparse scenarios. To address these challenges, we propose an end-to-end tensor completion framework tailored for financial data. It integrates an adaptive clustering module to characterize individual-specific heterogeneity, a regularized temporal smoothing module to preserve long-term dynamics while suppressing short-term noise, and a joint optimization strategy combining tensor decomposition with structural regularization for robust, structure-aware imputation. Extensive experiments demonstrate that our method consistently outperforms state-of-the-art baselines under diverse missingness mechanisms: asset pricing errors decrease by 12.6%–23.4%, and portfolio Sharpe ratios improve by an average of 18.9%. These results validate both the theoretical rigor and practical financial value of the proposed approach.

Technology Category

• Machine Learning: Matrix & Tensor Methods
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Game Theory and Economic Paradigms: Imperfect Information

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Missing data in financial panels presents a critical obstacle, undermining asset-pricing models and reducing the effectiveness of investment strategies. Such panels are often inherently multi-dimensional, spanning firms, time, and financial variables, which adds complexity to the imputation task. Conventional imputation methods often fail by flattening the data's multidimensional structure, struggling with heterogeneous missingness patterns, or overfitting in the face of extreme data sparsity. To address these limitations, we introduce an Adaptive, Cluster-based Temporal smoothing tensor completion framework (ACT-Tensor) tailored for severely and heterogeneously missing multi-dimensional financial data panels. ACT-Tensor incorporates two key innovations: a cluster-based completion module that captures cross-sectional heterogeneity by learning group-specific latent structures; and a temporal smoothing module that proactively removes short-lived noise while preserving slow-moving fundamental trends. Extensive experiments show that ACT-Tensor consistently outperforms state-of-the-art benchmarks in terms of imputation accuracy across a range of missing data regimes, including extreme sparsity scenarios. To assess its practical financial utility, we evaluate the imputed data with an asset-pricing pipeline tailored for tensor-structured financial data. Results show that ACT-Tensor not only reduces pricing errors but also significantly improves risk-adjusted returns of the constructed portfolio. These findings confirm that our method delivers highly accurate and informative imputations, offering substantial value for financial decision-making.

Problem

• Imputing missing multi-dimensional financial data panels
• Addressing heterogeneous missingness patterns and data sparsity
• Improving asset-pricing models and investment strategy effectiveness

Innovation

• Cluster-based completion for cross-sectional heterogeneity
• Temporal smoothing to remove noise and preserve trends
• Adaptive tensor framework for extreme sparsity scenarios