Project Details
Projekt
FOR 5381: Mathematical Statistics in the Information Age - Statistical Efficiency and Computational Tractability
Subject Area
Mathematics
Term
since 2021
Website
Homepage
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 460867398
State-of-the-art massive data analysis consists of consecutively running a series of individual processing steps on the data. In order to approach statistical efficiency in a computationally tractable way and to provide valid statistical inference for the composite data processing pipeline, the single processing steps cannot be studied isolated from each other. This is where this Research Unit takes its unique place in the scientific landscape. On a profound mathematical basis, our goal is the conjoint development of comprehensive statistically efficient and computationally tractable valid inference methodology for composite data processing. Through the development of a rigorous theoretical foundation, we aim to make a broad impact across all scientific domains of modern statistical data analysis.
DFG Programme
Research Units
International Connection
Austria
Projects
- Computationally tractable bootstrap for high-dimensional data (Applicants Dette, Holger ; Rohde, Angelika )
- Coordination Funds (Applicant Rohde, Angelika )
- Coordination Funds (Applicant Rohde, Angelika )
- Optimal actions and stopping in sequential learning (Applicants Carpentier, Alexandra ; Reiß, Markus )
- Preprocessed and high-dimensional data sets in discriminant analysis and classification (Applicants Langer, Sophie ; Rohde, Angelika )
- Sublinear time methods with statistical guarantees (Applicants Dette, Holger ; Munk, Axel )
- Supersmooth functional data analysis and PCA-preprocessing (Applicant Meister, Alexander )
Spokesperson
Professorin Dr. Angelika Rohde
