Research area Statistics

Statistics is the science of collection and analysis of data. The research field develops probability models and methods for data analysis and uncertainty quantification, often by a combination of mathematics and advanced computer algorithms.

Researchers in Statistics are developing inference methods for drawing statistically verified conclusions about underlying mechanisms from observed data. Data often contain errors of different kinds, and a central part of statistical research is therefore focused on data collection methods for improved data quality. Methods for predicting new data and decision making under uncertainty are also important parts of statistical research.

Statistical methods are used in all empirical sciences, and statistical researchers are therefore often active in specialized subareas such as econometrics, psychometrics and demography. Statistics is also fundamental for artificial intelligence and machine learning, areas that have become increasingly important for statistical research.

Bayesian Inference

Bayesian statistics is the formal process of updating beliefs about uncertain quantities in light of new data. The uncertainty is quantified by subjective probabilities, leading to a natural and practical way to make predictions and optimal decisions under uncertainty.

Go to Bayesian Inference

Design of experiments

Experiments allow the study of research questions under controlled settings, such that the effects of the examined variables can be isolated. The aim is to explore a possible relationship between one or more input/treatment variables and a response variable of interest, and to estimate the sizes of any treatment effects. With careful planning of the experiment, valid conclusions can be drawn, and the amount of information can be maximized given the available resources.

Go to Design of experiments

Linear Models and their Extensions

Statistical models are used to explain the variability of measurements made on the variable of interest (also called outcome, response).

Go to Linear Models and their Extensions

Massively dependent data

The standard paradigm for statistical modelling assumes that observations on variables are independent. In addition to multivariate statistical techniques that afford for dependencies between variables, and time series analysis with dependence over time, many empirical settings give rise to complex data structures that place additional requirements on modelling and inference.

Go to Massively dependent data

Official Statistics

The research area comprises all aspects of production of official statistics and similar types of statistics.

Go to Official Statistics

Statistical models in the social sciences

This theme is an umbrella for a number of research areas in statistical models and associated method, with applications (mostly) in the social sciences.

Go to Statistical models in the social sciences

Time series analysis

Time Series Analysis is the study of data collected sequentially over time, with the aim of understanding dependencies, extracting meaningful patterns, and making predictions about future observations. The field has its modern foundations in the work of Box and Jenkins (1970), and has since then been shaped through intensive research e.g. by the Nobel Prize-winning contributions on autoregressive conditional heteroskedasticity and cointegration (Engle and Granger, 2003) and vector autoregressions for causal inference (Sims, 2011).

Go to Time series analysis

Cookie settings

By allowing third-party services, you accept their cookies and the use of tracking technologies necessary for them to function properly.