Mapping Time Series to Networks
Author Information
Author(s): Campanharo Andriana S. L. O., Sirer M. Irmak, Malmgren R. Dean, Ramos Fernando M., Amaral Luís A. Nunes
Primary Institution: Instituto Nacional de Pesquisas Espaciais, São José dos Campos, São Paulo, Brazil
Hypothesis
Can network analysis be used to characterize time series and vice versa?
Conclusion
The proposed mapping retains much of the information from the original time series and allows for the analysis of both time series and networks.
Supporting Evidence
- The mapping retains much of the information encoded in the original time series.
- Distinct time series result in networks with different topological properties.
- The method preserves both structured and unstructured information in both domains.
Takeaway
This study shows how we can turn a series of data points into a network and then back again, helping us understand patterns in both.
Methodology
The study introduces a map from time series to networks and demonstrates its properties through simulations and comparisons of generated time series and networks.
Limitations
The mapping is not strictly one-to-one due to its stochastic nature.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website