Dr. Helmut Farbmacher
 
University of Mannheim



University of Mannheim; Spring Term 2018

Regression Shrinkage Methods

Seminar



Announcements

  • The introductory meeting is on Thursday (8.3.) from 3.30-5.00pm, P043, L7, 3-5.
  • The seminar presentations will be on Friday (4.5., 10am-5pm, P043, L7, 3-5) and Saturday (5.5., 9-12am, 003, L9, 1-2)

Term paper and grading

"Big Data" methods have become increasingly important in data analysis. The students should be enabled to understand basic concepts in regression shrinkage and to utilize recent results for their own applied work.

The seminar gives an introduction in regression shrinkage methods. Both econometric theory and (economic) applications will be included in the course.

Details of the course will be discussed and a short lecture will be given during the introductory meeting on Thursday (8.3. 15.30-17.00pm, P043, L7, 3-5). The seminar will be held in English. The seminar presentations will be on Friday (4.5., 10am-5pm, P043, L7, 3-5) and Saturday (5.5., 9-12am, 003, L9, 1-2).

The seminar paper must be substantially your own work. It must be written in English. To keep chances equal, I will not answer specific questions to your seminar paper. Submission is online via email in pdf-format.

For each topic you will find a short description of a proposed task. You can either focus on a single paper and present this paper in more detail (problem, solution of the problem, main results, intuition, sketch/idea of proof (if not too long or complicated), application, etc.) or write some kind of survey summarizing several papers.

Please do not simply copy the paper but present it in your own way, explaining the important concepts in own words. This will be a major part of your task. When you copy results or tables from other studies, try to focus on the main information, which is relevant for your argument. Do not simply copy and/or discuss all results.

Replications of a paper (e.g. with an own application and/or Monte Carlo simulations) are also welcome and appreciated, but not required.



21.02.2018 Send feedback to Helmut Farbmacher.