time series harmonization

This study addresses the limitations of existing hierarchical time series forecasting methods, which are predominantly univariate and struggle to simultaneously satisfy aggregation constraints and exploit inter-variable correlations. To overcome this, we propose a multivariate joint reconciliation framework that explicitly incorporates the correlation structure among variables into the reconciliation process—marking the first such approach to move beyond traditional univariate, independent reconciliation. Built upon a multivariate regression framework, our method integrates base forecasts with covariance information and achieves coherent predictions across both variables and hierarchy levels by minimizing a multivariate loss function. Empirical evaluations on both simulated data and real-world Brazilian employment statistics demonstrate that the proposed approach significantly outperforms state-of-the-art methods, yielding markedly improved forecast accuracy.

Existing time series chain methods are confined to individual sequences and struggle to capture anomalous evolution patterns across interruptions or related sequences. This work proposes a novel formulation—Joint Time Series Chain—that extends the concept of time series chains to cross-sequence scenarios for the first time. By leveraging subsequence similarity modeling, a cross-sequence alignment strategy, and a new chain-ranking criterion, the method efficiently uncovers robust anomalous evolution trends. Empirical evaluations demonstrate that the proposed approach significantly outperforms current state-of-the-art techniques across multiple datasets. Furthermore, it has been successfully deployed in an Intel manufacturing setting, where it effectively identifies complex anomalous patterns, showcasing its practical utility in real-world industrial applications.

Existing time-series visualization methods face a fundamental trade-off between scalability and visual clarity, struggling to simultaneously support efficient exploration and readability for long, multivariate time-series data. This paper introduces TiVy, the first algorithm that jointly leverages dynamic time warping (DTW) and frequent sequence pattern mining to achieve time-aligned, variable-length, non-overlapping subsequence clustering. By integrating DTW-based symbolic representation with an efficient grouping strategy, TiVy constructs lightweight visual summaries and implements an interactive framework enabling real-time rendering. Experiments demonstrate that TiVy accelerates DTW-based clustering by up to three orders of magnitude compared to conventional approaches, while significantly improving both pattern extraction accuracy and visualization interpretability. In two real-world applications, TiVy successfully uncovers latent structural patterns and semantic regularities in large-scale time-series datasets.

This work proposes an interactive visual analytics approach to address the challenges of visual clutter and redundancy in large-scale time series visualization, which often obscure critical trends. By integrating M4 sampling, dynamic time warping (DTW) similarity computation, and a greedy selection strategy, the method automatically identifies a representative subset of time series that preserves essential patterns while minimizing redundancy. A coordinated multi-view visualization framework further enables users to efficiently explore and interpret the data. The proposed technique significantly enhances visual clarity, interpretability, and analytical efficiency without sacrificing the core temporal characteristics of the original dataset.

This work addresses the critical need for unified analysis of graph and time series data, an area currently lacking systematic organization in existing systems. It proposes the first comprehensive taxonomy that categorizes fusion architectures into four distinct classes. Through a multidimensional evaluation grounded in cross-model integration depth, maturity, and openness, the study establishes a clear classification framework via literature review, architectural analysis, and requirement mapping. This framework delineates the appropriate application scenarios and inherent design trade-offs for each architecture type, thereby offering researchers and practitioners a principled guide for system selection and identifying promising directions for future research.

This work addresses the limitations of traditional time series methods, which are constrained by fixed forecasting horizons and struggle to support contextual reasoning, tool invocation, and structured decision-making in real-world scenarios. The authors propose AION, a framework that formalizes time series tasks as a triplet of task specification, workspace, and validation interface, integrating six core modules—agents, skills, rules, memory, evaluation, and protocols—to emphasize temporal grounding, knowledge-guided reasoning, and reliability assurance. By incorporating process traceability, multi-level auditing, and post-hoc experimental analysis, AION overcomes the constraints of static evaluation paradigms. In a Kaggle store sales forecasting case study, AION substantially outperforms direct modeling approaches, generating richer reasoning traces, intermediate artifacts, and audit steps, thereby demonstrating its effectiveness and superiority in handling complex, real-world time series tasks.

This study addresses the lack of systematic evaluation of stationarity-inducing transformations across diverse non-stationary time series. The authors construct synthetic datasets encompassing trend, seasonality, and heteroskedasticity, complemented by real-world airport passenger flow data, and conduct 3,528 controlled experiments evaluating 14 transformation methods across seven forecasting models and three prediction horizons. Innovatively, stationarity is assessed via consensus from ten statistical tests, and mediation analysis elucidates underlying mechanisms. Results challenge the common assumption that transformations universally improve forecasts: matched transformations enhance accuracy in only 18% of cases; log or Box–Cox transformations are effective for heteroskedastic data (60–65% of cases); and differencing consistently degrades performance on linear-trend series.

Pretrained time series foundation models often underperform on downstream tasks due to domain shift, task heterogeneity, scarce labeled data, and computational constraints. This work proposes the first systematic post-training framework, categorizing existing approaches along five dimensions based on their intervention points within the forecasting pipeline: parameter adaptation, context augmentation, model composition, output and uncertainty calibration, and compression with specialization. By delineating the design space and inherent limitations of each category, the framework offers a structured pathway to bridge the gap between pretraining and reliable deployment, thereby advancing the standardization and systematic development of time series post-training methodologies.

This work addresses the limitations of existing foundation time series models, which suffer from high computational overhead, poor adaptability to dynamic data streams, and an inability to learn continuously—hindering their deployment in resource-constrained environments. To overcome these challenges, we propose TimeBlocks, a novel foundation modeling paradigm that uniquely integrates multi-task generalization, lightweight architecture, and continual calibration capabilities. TimeBlocks dynamically assembles compact models at inference time through a modular pool of model blocks and a routing strategy tailored to incoming data streams. Furthermore, it incorporates StreamCore, a streaming summarization algorithm that enables efficient online calibration. Extensive experiments demonstrate that TimeBlocks achieves significantly higher prediction accuracy than current methods across multiple datasets while maintaining low computational costs, enabling effective real-time forecasting under stringent resource constraints.