Research in Statistics

The research at the department of statistics mainly concerns two fields:
econometrics/time series analysis and Structural Equations Modeling (SEM). Within econometrics/time series analysis, we study macro economic models, in particular analysis of long term relations, i.e. cointegration. If data from many countries are present, we obtain a panel. Panel cointegration is a relatively new field, which is of great interest to us.

We are also interested in analysis of financial data. Many empirical studies indicate that such data are volatile, which means that their degree of fluctuation differs between different time periods. (For example, the volatility is high during periods of financial crisis.) Hence, there is a great need for statistical methods that take this kind of behavior into account.

To a large extent, our SEM research amounts to factor analysis, which tries to make conclusions about latent (non observable) factors. For example, such factors may explain the outcome of a questionnaire. A useful tool is the program package LISREL, which is developed by researchers at our department. SEM finds many applications in behavioral sciences.