Time Series Analysis in Machine Learning
Time Series Analysis in Machine Learning
Authors: A. Pagliaro, A. Anzalone Date: 2026-06-10 Paper ID: openalex:2606.11746
Summary
This paper serves as a pedagogical overview of time series analysis, bridging the gap between classical statistical models and modern machine learning approaches. It details essential concepts such as stationarity and seasonality while examining the evolution from ARIMA and state-space models to advanced neural architectures like transformers and CNNs. By integrating examples from fields such as astronomy and finance, the authors provide a unified framework for researchers to select and apply appropriate temporal modeling techniques.
Key Contributions
- Provides a comprehensive pedagogical review linking classical statistical time series analysis with modern machine learning paradigms.
- Systematically categorizes time series techniques ranging from traditional autoregressive models to deep learning architectures like transformers and RNNs.
- Synthesizes common principles across interdisciplinary domains including astrophysics, meteorology, and finance to guide practical research applications.
Links
Metadata & Links
- url
- https://arxiv.org/abs/2606.11746
- paper_id
- 2606.11746
- paper_source
- openalex
- domain
- nlp
- tags
-
- architectures
-
- datasets
-
- concept_slugs
-
- dataset_slugs
-
- skill
- TimeSeriesSkill
- created_at
- 2026-06-13T06:27:54Z