Research in Statistics
Research by application areas
Brain Research and sensor data
 Independent component analysis
 Modelling eyemovement data
 Bayesian methods in diffusion magnetic resonance imaging of brain
 Costefficient design of experiments
Biological and environmental science
 Estimating the size and structure of moose populations
 Generalized linear latent variable models for joint modeling in ecology
 State space models
 Statistical method development in forestry
 Advanced Computational and Statistical Techniques for Biomonitoring and Aquatic Ecosystem Service Management
Population studies in health and human sciences
 Nonparticipation in health examination surveys
 Costefficient design of observational studies
 Causal models
 Statistical methods in sport and health sciences
Others
 Nonparametric and robust methods
 Fibre processes
 Statistical analysis of spatiotemporal point patterns
 Structural equation models
 Empirical mode decomposition (EMD)
Research by areas of statistics
The department has special expertise in the following areas of statistics
 Computational statistics and simulation methods
 Nonparametric and robust methods
 Mixed models
 Missing data
 Time series analysis
 Processes
 Structural equation models and causal inference
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 statespace models
 causaleffect: The package causaleffect can be used to derive expressions and transportability formulas for causal effects in semiMarkovian causal models.
 fICA: Algorithms for classical symmetric and deflationbased FastICA, reloaded deflationbased FastICA algorithm and an algorithm for adaptive deflationbased 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: Lmoments and quantile mixtures
 PearsonICA: Independent component analysis using score functions from the Pearson system
 rankhazard: Rankhazard 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 nonstationary and nonlinear 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.