Carsten Trenkler is Professor of Empirical Economics at the University of Mannheim since 2007. He also holds a position as Research Professor at the Institute for Employment Research (IAB) in Nuremburg. He received his doctoral degree in Economics in 2002 from the Humboldt University (Berlin). He subsequently worked in various research and teaching positions at the universities of Florence (Italy), Bonn, Berlin and Mannheim.
Carsten's main research interests are:
Data and Methods
Econometrics is about analyzing economic data using statistical techniques and economic theory. With the help of econometric methods one may answer many interesting questions such as: Does participation in a job training program reduce unemployment duration? How much do tobacco taxes reduce smoking? Does reducing class size improve performance of pupils? How large is this effect? Is the size of this effect uniform or rather depends on the context? How much are sales increased by advertisement? What will the inflation rate be next year? How much does a decrease in the European Central Bank’s interest rate affect growth in the gross domestic product in the next two years? No doubt, econometrics is a very important branch of economics, finance as well as business economics. Econometric methods are also used in other social sciences like political science or sociology.
There are two major reasons for analyzing economic data: (i) economic models often do not give the quantitative information that is needed in practical decision making and (ii) there are competing models making different predictions. Consider e.g. the introduction of a minimum wage. Depending on the assumptions, economic models may suggest different qualitative effects of the introduction of a minimum wage on employment in the economy. Using adequate data and econometric methods may help to measure the direction and the magnitude of the effect of a minimum wage on employment.
In empirical analyses one would like to identify so-called causal effects rather than just measuring correlation between variables. If you collect data on the density of individual’s head of hair and their wealth, you would probably figure out that the less hair of head they have, the richer they are. Shall you recommend to have one’s hair cut and going bald in order to become rich? You should only do this if there is a causal link from hair density to wealth. We would not expect this link to be present. Rather, there is simply correlation between hair density and wealth because, on average, older people are richer and have less hair.
It is relatively easy to measure causal effects if data are generated by an experiment. However, in economics we often just have observational data at hand that have been collected e.g. by statistical offices or have been obtained via surveys. Why is that a problem? If one would like to measure the causal effect of the class size on pupil’s test scores, one prefers to only vary the class size while keeping all other potential influences on test scores constant. You can imagine that it is impossible to implement such a perfect experiment.
Although, it has been tried to implement randomized versions of such class-size experiments, there may only be observational data on test scores, class size and a few other variables available. Hence, many other things may have changed besides the class size. There are, however, specific econometric approaches that can be used to try to identify a potential causal effect of the class size from such observational data.
We offer many econometrics courses in our Bachelor program. They introduce various econometric methods as well as software packages and show how to apply the methods to economic data. In particular, there are seminars and other courses in which students can run own empirical projects. If you would like to know a bit more about econometrics, you could have a look at the first chapter of two nice introductory textbooks: Introduction to Econometrics by J.H. Stock and M.W. Watson and Introductory Econometrics by J. M. Wooldridge.
There are currently no vacancies at our Chair.