While controlled randomized experiments provide data of a kind that otherwise would be unavailable for economists, the econometric analysis of the data still poses challenges. Subjects interact, usually repeatedly, and learn during the experiment. This introduces dependencies and correlations that have to be dealt with in the econometric analysis, as with other kinds of economic data. One important difference though is that the structure of the data depends on the design of the experiment. The topic of the course is thus both on the econometric analysis of experimental data, but also, with the analysis in mind, how to design the experiment to make the data most useful to answer your research question.
The course aims to be accessible to doctoral students who have taken only a first-year graduate course in econometrics. A student who has taken several econometrics courses beyond that will probably find most techniques familiar. A good deal of the course will be based on recent published papers in experimental economics and review and critically discuss the statistical techniques used so there is a chance that the student might find the course helpful nevertheless. If the students, however, expects a lecture-style course of advanced econometric methods, this one is probably not the best choice.
The course will focus mostly on the statistical and econometric analysis of data from experiments and devote relatively little time on the optimal design of experiments. So if a student is mostly interested in the latter, this won't be the right course for him/her. A good deal of the course will be based on recent published papers in experimental economics and review and critically discuss the statistical techniques used.