Time series, spatial statistics and signal processing

This theme covers research in the broad area of time series, spatial statistics and signal processing.

More specifically, the group pursues research on prediction theory and model selection for time series, wavelet methods in time series and point processes, graphical modelling of time series and analysis of the nature of complex-valued time series, time series generated by chaotic maps, continuous time modelling of time series (focussing in particular on high-frequency data) and statistical methods for stochastic processes, including long memory processes.

Moreover, we have expertise in point processes and spatial point patterns and in statistical image processing and statistical analysis of data collected over two and higher dimensional lattices with emphasis on Gauss-Markov random fields and on ambit fields.

Application areas covered by this group range from biomedical settings, transport modelling,  environmental variables to financial data and the modelling of energy markets.