Causal analysis of measured multivariate time series

PhD study program: Measurement Technology
Akademic year: 2022-2023
Advisor: RNDr. Anna Krakovská, CSc. (krakovska@savba.sk)
External educational institution: Institute of Measurement Science SAS
Accepting university: Faculty of Electrical Engineering and Information Technology, Slovak University of Technology in Bratislava, Institute of Electrical Engineering

Annotation:

The topic concerns networks in which individual nodes are characterized by measured time series. The emphasis is on detecting causal relationships between nodes. The main areas of application are multichannel electroencephalographic recordings from the human brain and multi-lead ECG measurements. One of the goals of multivariate causal analysis will be to find the variables with the strongest impact. In practice, when measuring multiple manifestations of a system, it can help to select the measurements that represent the most useful source of information. The thesis topic is suitable for a candidate with an interest in developing appropriate mathematical approaches and possibly also the creative application of machine learning methods. Good English skills and experience in creating and testing software in the MatLab environment are also a necessary requirement. As part of the doctoral study, the student will expand his/her knowledge in the field of biomeasurement and get acquainted with methods from the theory of dynamical systems, including chaos and fractal theory and partly on statistics, information theory and mathematical optimization.