arXiv:2607. 00154v1 Announce Type: cross Abstract: Evolutionary neural architecture design for multivariate time-series forecasting remains underexplored, with most approaches relying on fixed Transformer architectures despite substantial variation across tasks and forecasting settings.
arXiv:2602. 02288v3 Announce Type: replace Abstract: Current time-series forecasting models are primarily based on transformer-style neural networks.
arXiv:2510. 04487v5 Announce Type: replace Abstract: While accuracy is a critical requirement for time series forecasting, an equally important desideratum is reasonable forecast volatility across forecast creation dates (FCDs).
arXiv:2606. 27282v1 Announce Type: new Abstract: Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation models, on the assumption that capacity is what unlocks accuracy.
Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation models, on the assumption that capacity is what unlocks accuracy. We take the opposite position: most of the gap can be closed at far lower cost by tuning preprocessing rather than scaling models.
arXiv:2512. 22702v2 Announce Type: replace Abstract: Deep learning models have grown popular in time series applications.
The Nixtlaverse is an open‑source Python ecosystem that unifies statistical, machine‑learning, and neural forecasting models by sharing a common long‑format panel data structure and keyed forecast outputs while preserving each model family’s specialized implementation. Using the M5 competition data, the authors demonstrate the framework’s ability to evaluate diverse models in a single rolling‑origin setup, profile runtime and memory usage across thousands of series, and reconcile forecasts from multiple engines, including an external one, over the full M5 hierarchy. The paper highlights the design’s cost, boundary, and utility advantages and notes its growing adoption and permissive licensing.
Halo is a modification to existing deep forecasters that adds a second output for estimating the scale of the predicted distribution, trained with a matching negative log likelihood. Experiments on five electricity price markets show that Halo improves mean squared error and mean absolute error in 28 of 30 model‑market‑metric comparisons, with average MSE reductions of 2.6% to 16.5% and MAE reductions of 1.7% to 11.0%. The study finds that the source of the scale estimate is less important than the fact that the network estimates scale, and that the improvement persists without retuning hyperparameters.
arXiv:2603. 15506v2 Announce Type: replace-cross Abstract: We argue that the current practice of evaluating AI/ML time-series forecasting models, predominantly on benchmarks characterized by strong, persistent periodicities and seasonalities, obscures real progress by overlooking the performance of efficient classical methods.
The paper presents an empirical benchmark of nine modern deep‑learning models for time‑series forecasting of smart‑meter energy consumption, evaluated on two publicly available datasets. It examines how historical input length, prediction horizon, and model architecture affect accuracy, finding that longer historical context improves performance up to a saturation point and that accuracy declines with longer horizons. The study also compares computational complexity, showing that lightweight architectures achieve similar performance to heavier models, and notes that model choice has limited impact across most demographic and household subgroups.
The paper presents a comparative study of six deep learning models—state-space, MLP, RNN, and Transformer-based architectures—for cross-border electricity price forecasting using publicly available data. It focuses on generalization across markets and evaluates performance under low-data target-market conditions (zero-shot, one-shot, few-shot) with a standardized dataset for the Germany‑Luxembourg bidding zone in 2024. Results show that N‑HiTS and NBEATSx perform competitively in limited‑data scenarios, while transformer models achieve comparable accuracy but require more adaptation and tuning, and that careful feature selection and hyperparameter tuning improve performance.
Large forecasting applications often combine statistical, machine-learning, and neural models. These families solve the same problem but differ in fitted state, training procedures, and how they paral...