Answers are grounded in this paper and cite their sources.
Purpose. To conduct a structured analysis and classification of modern methods and models used to work with time series of various nature. Attention was paid not only to typical features and types of calculations, but also to identifying the subject area of application, comparing and highlighting strengths and weaknesses when working with different data sets, with relevant examples of areas of use and an emphasis on advantages. Methodology. A step-by-step and detailed review of existing methods and models based on their main characteristics, areas of use, and features of working with approaches of different nature that use different properties of time series. Findings. Analysis of the most common methods for processing time series and a separate review of their representatives. Particular attention is paid to hybrid models that can combine methods of one or different classes, as well as atypical approaches based on the specific properties of time series, in particular their fractality. Originality. It consists in a comprehensive and fundamental consideration of methods for analyzing time series, ranging from classical linear and nonlinear statistical models and artificial intelligence methods to hybrid and fractal approaches, with an emphasis on identifying their areas of application and comparing their advantages and disadvantages. The practical value of the research lies in the systematization of material that can be used for preliminary analysis of the subject area and selection of tools based on their effectiveness, which, in turn, simplifies the search for analogues and reduces the time required to prepare for research. In addition, the work highlights lesser-known and atypical methods that are of interest for further research and may be promising candidates for future scientific developments in the field of time series analysis.
No takes yet. Share an insight, caveat, or question.
Zhadan et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: