PhD programme

Causal inference aims to deepen the understanding of social phenomena, or of analyses of efficiencies of different treatments (such as, medicinal, work-life or environmental treatments). Statistics cannot by itself create knowledge of various phenomena, but is rather used to test theories under various assumptions. Thematic theory, together with statistical theory and knowledge of how data is collected, jointly form the causal analysis. For that reason, it is necessary to understand thematic questions, and that in the development of new methods jointly consider how data together with statistical theory can improve analysis of causal questions, such as testing of theories as well as pure effect evaluations.

Currently, we are working with researchers in medicine, economics, psychology and engineering. Some of our thematic work include: (i) test of theories for how mass media affect voters' ability to hold elected officials accountable, (ii) analysis of the effect of family friendly workplaces on wages and income for men and women, (iii) analysis of the effect of air pollution on childrens' health, (iv) test of gender differences in preferences and (v) analysis of electricity consumption and changes in electricity tariffs and consumer information related to energy savings. Methodologically, we have made contributions to, among others, the design of randomized experiments and identification of causal effects using observational data, and register data in particular. Examples of problems we have investigated concern situations when the timing of a treatment is a choice (as opposed to a randomized experiment when the time for intervention is the same for all treated and untreated individuals) and where there might exist measurement errors, both in control variables and in the timing of treatment.

Responsible researchers: Per Johansson and Ingeborg Waernbaum

Structural equation modeling (SEM) is a multivariate statistical analysis technique that simultaneously unites Factor Analysis and Multiple Regression Analysis. It analyses the causal relationships among observed variables and latent constructs, including linear and nonlinear effects. SEM includes two basic types of models. The measurement modelrepresents the theory that specifies how a set of observed variables measure the latent constructs. The structural model represents the theory that shows how latent constructs are causally related to each other.

SEM can apply to various data types, cross-sectional data, longitudinal data, time-series data, or multilevel data. For example, the cross-sectional models help us to assess causal and mediation hypotheses; Latent Growth Curve Models often apply to analyse potential changes in the latent construct of interest over time; Item Response Theory Models usually analyse the patterns of individual behaviours and questionnaire responses, and Multilevel Models can assess the cause of variations between different data levels. SEM has been widely applied in social sciences and spread to other natural sciences in recent decades, e.g., information science and medical research.

The Department of Statistics in Uppsala has a long tradition of structural equation modeling, and it is known as the birthplace of SEM. Professor Emeritus Karl G. Jöreskog is the pioneer in SEM, and the LISREL (linear structural relations) program (Jöreskog and Sörbom) was the first software for analysis of Structural Equation Models. Nowadays, Professor Fan Wallentin and Associate Professor Shaobo Jin with colleagues carry on this rich tradition and actively contribute to the field.

Responsible researcher: Fan Yang Wallentin

The Department of Statistics has a long tradition of research within times series econometrics. The tradition started in 1942 when Herman Wold become professor at the department. Econometrics is the field concerning applications of statistical methods on problems within economics, and the methods are developed for applied problems in economics. Time series econometric deal with economic data sorted over time, and the aim is to analyse associations between different economic variables and their development over time. For instance, the government need to know how the gross domestic product (GDP), inflation and unemployment will develop in order to make the national budget.

One research area is to improve forecast to just these variables (GDP, inflation and unemployment). We are living in a ever-changing world affected by various decision. These decisions result in associations between economic variables changing over time. Another research area is how to model structural changes over time using non-linear models. With such models, it is possible to get increased understanding of various economic phenomena.

Responsible researcher: Johan Lyhagen