Chair of Information and Coding Theory

etit-623: Time Series Analysis


"Time series analysis" represents a field of statistical approaches to analysing time-resolved data sets. In this lecture a first introduction into this field will be provided, including examples of applications. A certain focus will be put on data sets with biomedical or neuroscience background. There will be some overlap with the lecture "Applied Signal Processing I - Statistical Signal Processing". Tentative table of contents:

  • data processing in biological networks
  • acquisition of time series in neuroscience
  • inverse problem of brain source analysis
  • sets of linear equations / pseudoinverse
  • principal component analysis / independent component analysis
  • blind signal separation
  • maximum-likelihood estimation
  • entropy and mutual information
  • prediction of time series
  • whitening of residuals
  • dynamical systems
  • stochastic differential equations
  • deterministic chaos
  • autoregressive modelling
  • state space modelling
  • reconstruction of strange attractors
  • fractal dimensions / Lyapunov exponents




Recommended literature

  • J.D.Hamilton: "Time Series Analysis"
  • H.Kantz & T.Schreiber: "Nonlinear Time Series Analysis"