17.03.2025
| Conference |
Advancing Time Series Preprocessing: DSC at the 17th German Probability and Statistics Days
DSC research associate Maryam Movahedifar presented a talk at GPSD 2025 on preprocessing noisy time series using (M)SSA. Her contribution highlighted DSC's methodological work and will be submitted as a proceeding to the same conference she attended in Dresden.
From March 11–14, 2025, the 17th German Probability and Statistics Days (GPSD) took place at TU Dresden. As one of the most important events for the statistics and probability community in Germany and Europe, GPSD 2025 brought together researchers from a wide range of disciplines and career stages to discuss new developments, methods, and applications in the field.
Maryam Movahedifar, data scientist at the University of Bremen's Data Science Center (DSC), represented the DSC community by giving a talk during the conference. Her presentation focused on the often-overlooked but vital task of data preprocessing-particularly in the context of noisy, high-dimensional time series data.
While technical in nature, her work contributes to one of the core missions of the DSC: building robust, interdisciplinary methodologies that can be applied across domains. In her talk, Maryam presented a comparative study evaluating different preprocessing methods, with a focus on a nonparametric technique called (Multivariate) Singular Spectrum Analysis ((M)SSA). The aim was to assess how preprocessing quality affects the reliability of statistical analyses, especially in real-world, noisy datasets.
By sharing this research with the GPSD community, Maryam not only contributed to academic exchange but also raised the profile of the DSC's applied and methodological expertise. Her session sparked interesting discussions among participants working on statistical inference, time series analysis, and data-driven modeling.
Maryam is now preparing a written version of her contribution to be included in the official GPSD 2025 proceedings. This publication will detail the key findings of the project and offer guidance for other researchers facing similar challenges in their data.
This participation reflects the DSC's commitment to supporting its members in disseminating their work and engaging with the broader research community. Whether through methodological innovation or interdisciplinary collaboration, contributions like Maryam's help ensure the DSC remains a hub for cutting-edge data science in Bremen and beyond.
Updated by: Dr. Maryam Movahedifar
« back
17.03.2025 | Conference
Advancing Time Series Preprocessing: DSC at the 17th German Probability and Statistics Days
DSC research associate Maryam Movahedifar presented a talk at GPSD 2025 on preprocessing noisy time series using (M)SSA. Her contribution highlighted DSC's methodological work and will be submitted as a proceeding to the same conference she attended in Dresden.
From March 11–14, 2025, the 17th German Probability and Statistics Days (GPSD) took place at TU Dresden. As one of the most important events for the statistics and probability community in Germany and Europe, GPSD 2025 brought together researchers from a wide range of disciplines and career stages to discuss new developments, methods, and applications in the field.
Maryam Movahedifar, data scientist at the University of Bremen's Data Science Center (DSC), represented the DSC community by giving a talk during the conference. Her presentation focused on the often-overlooked but vital task of data preprocessing-particularly in the context of noisy, high-dimensional time series data.
While technical in nature, her work contributes to one of the core missions of the DSC: building robust, interdisciplinary methodologies that can be applied across domains. In her talk, Maryam presented a comparative study evaluating different preprocessing methods, with a focus on a nonparametric technique called (Multivariate) Singular Spectrum Analysis ((M)SSA). The aim was to assess how preprocessing quality affects the reliability of statistical analyses, especially in real-world, noisy datasets.
By sharing this research with the GPSD community, Maryam not only contributed to academic exchange but also raised the profile of the DSC's applied and methodological expertise. Her session sparked interesting discussions among participants working on statistical inference, time series analysis, and data-driven modeling.
Maryam is now preparing a written version of her contribution to be included in the official GPSD 2025 proceedings. This publication will detail the key findings of the project and offer guidance for other researchers facing similar challenges in their data.
This participation reflects the DSC's commitment to supporting its members in disseminating their work and engaging with the broader research community. Whether through methodological innovation or interdisciplinary collaboration, contributions like Maryam's help ensure the DSC remains a hub for cutting-edge data science in Bremen and beyond.
Author: Dr. Maryam Movahedifar
Please contact us if you have any questions:
Dr. Maryam Movahedifar
DSC Data Scientist
+49 (421) 218 - 59854
movahedm@uni-bremen.de
« back