Empirical Research in Energy Markets

The course offers: 1) A short recap / overview of basics in statistics. 2) An extended overview of regression modelling, including multiple regression, multivariate models, regression analytics. 3) Endogeneity considerations and other emprirical research obstacles, including possible solutions (e.g. instrument variable regression). 4) Principal component analysis, backgrounds and modelling 5) Data preprocessing and data transformation. 6) Overview of some practical studies from the field of Energy Markets. 7) Some basic domain knowledge from the area of Energy Markets. 8) Usage of statistical software ("R") to perform the studies.

Upon successful completion of this module, students will be able to understand and appraise the contents of empirical research papers. Furthermore, they are able to tackle empirical research obstacles. In addition, they are able to combine and jointly apply statistical concepts such as principal component analysis, multiple regression and instrument variable regression. Moreover, students are able to apply these concepts and methods in practice using the statistical software "R". Additionally, students can perform the preprocessing of data required before conducting research. Students are furthermore able to evaluate the usefulness of a variety of extension packages for "R", install and successfully use them in their research. Moreover, students will be able to apply the presented methods for their own research, in particular within the context of project study or Master’s thesis. Lastly, students know how to conduct their own studies and compose an overview of the findings.

The grading is based on a final written exam at the end of the term. The final exam will last 120 minutes. As an auxiliary, only a non-programmable calculator is allowed. The exam includes MC- questions, aiming at assessing the understanding of a variety of basic statistical concepts and formal concepts. Apart from that, the MC-questions also aim at evaluating the ability of students to choose the best strategy to overcome the typical obstacles of empirical research. Moreover, students are asked to enumerate some notable and prominent findings of empirical research in Energy Markets. Besides that, students also have to answer some open questions, where they can prove a deeper understanding of the Empirical Research methods by applying these to new problems. Lastly, some technical / mathematical questions and simple proofs can also be a small part of the exam, assessing whether the students are able to work with presented mathematical tools on their own. As mid-term, students who complete an exercise sheet and reach at least 80% of the points can get a grade bonus which improves the final grade by one notch (i.e. from 2.0 to 1.7). By completing the mid-term exercises, students can apply empirical research methods to real problems based on data sets from Energy Markets area. Further, students can evaluate which methods are most effective depending on the research question.

Knowledge of basic statistics is recommended (e.g. WI000258). Other modules of the Center for Energy Markets are recommended as well, but aren't a strong prerequisite. (e.g. WI000946)

In order to achieve the learning outcomes, the first part of the lecture consists of a presentation, along with some interactive elements, where students can actively participate by expressing their opinions or contributing with their prior knowledge. The slides-presentation are supplemented by calculations on the whiteboard allowing students to comprehend the derivation of statistical results. Students are be encouraged to deepen their understanding of the topics by reading the recommended literature and other reference materials. This allows students to get an overview of some relevant empirical research results as a part of self-study. Apart from that, students are introduced to the common obstacles of empirical research and possible solutions to them. In order to practice this knowledge, students are performing practical computer exercises using the software "R". In these exercises, students start with a raw data set and perform all the steps required to answer a particular research question, including data preprocessing, data transformation and others. As a mid-term, students are further encouraged to apply their skills by answering practical research questions using "R" and interpreting the results. Submitting the results in written form allows students to practice writing and presenting of statistical findings.