ACT-Tensor: Tensor Completion Framework for Financial Dataset Imputation

TL;DR

The paper tackles missing data in multi-dimensional financial panels used for asset pricing by introducing ACT-Tensor, a tensor completion framework that preserves cross-sectional heterogeneity and temporal dynamics. It combines a cluster-based completion module with a temporal smoothing module to combat extreme sparsity and non-stationarity, and validates the approach with a tensor-based asset-pricing pipeline. Results show ACT-Tensor achieves superior imputation accuracy across regimes and translates this into markedly improved pricing accuracy, factor signals, and risk-adjusted returns. The framework demonstrates robustness under structured missingness and provides a practically valuable tool for financial decision-making.

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.

Figures (1)

• Figure 1: RMSE Sensitivity under (a) Block and (b) Logistic Missing Regimes Against the Regularization Coefficient $\lambda$.