event correlation

Existing methods struggle to robustly quantify dependencies between continuous time series and discrete event sequences, often suffering from quantization errors, repeated values, and event redundancy. This work proposes the first non-parametric mutual information estimation framework that operates without data transformation or learning, directly modeling the continuous–discrete duality. It further introduces latent event clustering to mitigate co-occurrence bias. Evaluated on both synthetic and real-world datasets, the method consistently outperforms state-of-the-art approaches across tasks including causal analysis, recurring pattern discovery, and covariate selection, achieving notable improvements in accuracy, robustness, and interpretability.

This study investigates the sensitivity of causal discovery methods to mismatches between observed sampling times and true event times. By systematically analyzing how varying sampling rates and observation window lengths affect causal inference performance—and integrating insights from signal processing theory—the work reveals, for the first time, distinct stability characteristics between classical and modern causal discovery algorithms under changes in sampling hyperparameters. Combining theoretical analysis with empirical evaluation, the research demonstrates that the performance of several mainstream methods is highly dependent on sampling configurations. These findings provide novel theoretical foundations and practical guidance for temporal causal modeling, highlighting the critical role of sampling design in reliable causal inference from time-series data.

To address the dual challenges of time-varying causal discovery and future-value forecasting in multivariate co-evolving data streams, this paper proposes ModePlait—the first streaming incremental learning framework that unifies causal discovery and prediction. Its core contributions are: (1) an adaptive dynamic mode transition detection mechanism that accurately captures nonstationary, time-varying causal structures; and (2) a synergistic integration of adaptive sliding windows with lightweight parameter updates, enabling efficient, scalable stream processing over unbounded data. Evaluated on both synthetic and real-world datasets, ModePlait outperforms state-of-the-art methods by improving causal structure identification accuracy by 12.6% and reducing multi-step prediction error by 9.3%, thereby achieving a superior balance among accuracy, computational efficiency, and dynamic adaptability.

To address the trade-off between computational efficiency and estimation accuracy in time-series causal inference, this paper proposes Pairwise Edge Measures (PEMs) grounded in process motifs—introducing process motif theory to lagged covariance modeling for the first time. The proposed PEMs comprise two variants, each analytically correcting for confounding and reverse causation, respectively. The method provides rigorous theoretical guarantees for linear stochastic processes and admits a lightweight implementation requiring only the lagged correlation matrix. In extensive simulations, PEMs match or exceed the accuracy of Granger causality, transfer entropy, and convergent cross-mapping, while reducing computational time substantially—particularly advantageous for large-scale linear systems.

Causal discovery from binary alarm sequences in large-scale systems remains challenging due to the joint requirements of computational efficiency, sparse dependency modeling, and semantic capture of state transitions. Method: This paper proposes the first causal inference framework specifically designed for binary anomaly data. It introduces a sparse causal testing mechanism based on an improved Granger causality test, integrating flag-sequence feature encoding, adaptive graph-structure learning, and dynamic edge pruning. Contribution/Results: The framework explicitly models both the state-transition semantics and extreme sparsity inherent in binary data—novelty not addressed by prior work. By combining link compression with accuracy-aware pruning, it achieves scalable yet precise causal discovery. Evaluated on the CMS detector readout box system and IT monitoring datasets, it significantly reduces computational overhead while improving causal F1-score by a medium margin, thereby enabling effective real-time root-cause diagnosis.

This study addresses the challenge of accurately identifying directional lead-lag relationships among variables in multivariate time series. It proposes a novel framework grounded in temporal irreversibility, rigorously defining directionality as behavioral asymmetry under time reversal. Central to this approach is the introduction of a circulation matrix derived from lagged covariances as the core measure of directionality. The methodology incorporates an unbiased estimator, cross-fitting debiasing, delete-block jackknife standard errors, and an exact randomization test under a block-wise null hypothesis, with extensions to nonlinear settings. Evaluated on four simulated systems with known directional structures, the proposed method substantially outperforms conventional approaches—including correlation networks, Granger causality, and transfer entropy—demonstrating markedly reduced misidentification of directional relationships.

This study addresses the limitations of existing causal discovery methods, which rely on regular sampling and fixed lag structures and thus struggle with irregularly sampled time series commonly found in sensor, medical, and financial domains. To overcome this challenge, the work extends the PCMCI+ framework—originally designed for regularly spaced time series based on conditional independence tests—to irregular event streams by introducing a time-window aggregation mechanism. This mechanism replaces conventional fixed-lag modeling and enables time-aware causal inference. Empirical evaluations demonstrate that the proposed approach accurately recovers ground-truth causal graphs across synthetic irregular datasets under varying signal-to-noise ratios, significantly outperforming the original PCMCI+ and effectively eliminating the dependency on regular temporal structure.