Department of Mathematics and Statistics

Research in Statistics

Research in statistics in Department of Mathematics and Statistics contains both theoretical and application oriented research. Motivated by problems in applied fields, statisticians develop new efficient estimation methods and study designs. The research topics are listed below according to application areas. The list is dynamic: statisticians are open for new collaborative projects.

Research by application areas

Brain Research and sensor data

Biological and environmental science

Population studies in health and human sciences

Others

Research by areas of statistics

The department has special expertise in the following areas of statistics

Software

The department has expertise also on statistical software (R, SAS, SPSS, BUGS/JAGS).
The following R packages have been created at the department:

  • BSSasymp includes functions to compute the asymptotic covariance matrices of mixing and unmixing matrix estimates of several classical blind source separation (BSS) methods.
  • bssm: Bayesian inference of state-space models
  • causaleffect: The package causaleffect can be used to derive expressions and transportability formulas for causal effects in semi-Markovian causal models.
  • fICA: Algorithms for classical symmetric and deflation-based FastICA, reloaded deflation-based FastICA algorithm and an algorithm for adaptive deflation-based FastICA using multiple nonlinearities.
  • JADE: Cardoso's JADE algorithm as well as his functions for joint diagonalization are ported to R. Also several other blind source separation (BSS) methods, like AMUSE and SOBI, and some criteria for performance evaluation of BSS algorithms, are given.
  • KFAS: Tools for modelling multivariate exponential family state space models such as structural time series, ARIMA models, generalized linear models and generalized linear mixed models.
  • Lmoments: L-moments and quantile mixtures
  • PearsonICA: Independent component analysis using score functions from the Pearson system
  • rankhazard: Rank-hazard plots to visualize the relative importance of covariates in a proportional hazards model
  • Rlibeemd: An R package for performing the ensemble empirical mode decomposition (EEMD), its complete variant (CEEMDAN) or the regular empirical mode decomposition (EMD) for non-stationary and non-linear time series.
  • SpatialNP: The package contains test and estimates of location, tests of independence, tests of sphericity and several estimates of shape all based on spatial signs, symmetrized signs, ranks and signed ranks.
  • spatialsegregation,
  • smatr: The package provides (robust) methods of fitting bivariate lines in allometry using the major axis (MA) or standardised major axis (SMA), and for making inferences about such lines.
  • seqHMM: Fitting, evaluating, and visualizing hidden Markov models and complex sequence data.
  • tsPI: Package tsPI computes prediction intervals for ARIMA and structural time series models by using importance sampling approach with uninformative priors for model parameters, leading to more accurate coverage probabilities in frequentist sense.
  • tssBSS: provides algorithms to solve the blind source separation problem for time series with stochastic volatility.