What question did this study set out to answer?

To develop a robust, efficient method for estimating correlation matrices in high-dimensional data streams.

January 6, 2026MathematicsOpen Access

FORCE: Fast Outlier-Robust Correlation Estimation via Streaming Quantile Approximation for High-Dimensional Data Streams

View Full Paper

Ask AI

Bookmark

Authors

Streaming algorithm improves correlation estimation speed by 470× with robust performance in contaminated data.

To develop a robust, efficient method for estimating correlation matrices in high-dimensional data streams.
Introduced FORCE, a streaming algorithm using adaptive trimming via the P2 algorithm.
Compared performance of FORCE with six baseline algorithms including TP-Exact and FastMCD.
Evaluated across five benchmark datasets from various domains.
FORCE achieves speedups of approximately 470× over FastMCD and 3.9× over Spearman’s rank correlation.
On S&P 500 financial data, TP-Exact provided the best RMSE of 0.0902 compared to 0.1186 for FORCE.
FORCE maintains 76% accuracy of TP-Exact while using 104× less memory.

Jang et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark