AØKK08333U Seminar: Bayesian Econometrics
This seminar will allow students to gain practical experience with Bayesian econometric methods.
Bayesian methods offer a fresh perspective to econometrics, as they allow to tackle complicated estimation problems in a tractable way. These approaches rely on simulation methods and can therefore have an advantage over classical methods. For instance, unobserved variables (like latent utilities or random effects) may be difficult to integrate out of a likelihood function in a classical way, but are generally straightforward to simulate in a Bayesian framework.
Students will be able to choose the topic they want to explore, which can be either empirical or theoretical. It will also be possible to replicate the results of an article published in a scientific journal. To be able to tackle more interesting research questions, work in groups of two students will be highly encouraged.
Additional for the learning outcome specified in the Curriculum, the learning outcome of the seminar is to be able to
- Have reviewed the relevant literature related to the topic they have chosen, and understand the state of the art as well as the limitations of the current approaches.
- Have a grasp of simulation methods, understand their principle and how they can be used to make inference.
- Demonstrate an ability to select the most appropriate method for the topic they have chosen.
- Be able to implement Markov chain Monte Carlo methods, both theoretically (analytical derivation of the algorithm) and practically (programming).
- Demonstrate technical skills in writing code to implement Bayesian methods.
Conduct a full Bayesian analysis: (1) formulate an economic model, (2) organize prior knowledge about the model (prior), (3) use relevant data to express the observed information in the model (likelihood), (4) use Bayes' theorem to update beliefs (posterior), (5) derive an appropriate algorithm to compute the posterior distribution, (6) write code to implement the algorithm, (7) interpret the results and criticize the model.
- Lynch, Scott M. (2007). IntroductiontoAppliedBayesianStatisticsandEstimationforSocial Scientists. Springer. ISBN 978-0-387-71264-2.
- Lancaster, Tony (2004). AnIntroductiontoModernBayesianEconometrics. Blackwell Publishing. ISBN 978-1-405-11720-3.
- Other references and scientific articles will be suggested
to the students based on the subject they decide to study.
The student should have some programming skills (e.g., MATLAB, or the R programming language, which provides many freely available packages implementing Bayesian methods, Python, etc). A list of tutorials will be provided before the start of the semester for those who need to refresh their programming skills.
Before the session a "so-finalized-as-possible"-version of the paper must be uploaded in Absalon. After the presentations, the student submit an edited version of the paper in the Digital Exam portal as the final exam paper. The aim is that students use the presentation sessions as an opportunity to receive and use the constructive feedback to improve the paper.
• Kick-off meeting: September 3th 2018, 13-15 (General introduction to the seminar, discussion of possible topics.)
• Deadline commitmentpaper: September 21th, not later than 5 p.m. Before that you can arrange a meeting with the supervisor to discuss topics and related literature.
• Midterm poster session with feedback: in week 41 (to be determined at kick-
• Deadline of pre-paper uploadet to Absalon: One week before presentations
• Presentations/Workshops: Week 46 or 47 (dates to be determined at the kick-off meeting).
Read about the study programme and curricula at MSc in Economics
- 7,5 ECTS
- Type of assessment
- Written assignment- a seminar paper in English that meets the formal requirements for written papers stated in the curriculum of the Master programme and at KUNet for seminars.
- Exam registration requirements
Attendance in all activities at the seminar as stated in the formal requrements in the Curriculum and at the KUnet for seminars (UK) and KUnet for seminars (DK) is required to participate in the exam.
- All aids allowed
- Marking scale
- 7-point grading scale
- Censorship form
- External censorship
- Exam period
Autumn semester 2018:
Deadline for submitting the final seminar paper in DE: November 30, 2018 before 10.00 a.m.
The reexam is a written paper as stated in the Master curriculum.
Criteria for exam assesment
Students are assessed on the extent to which they master the learning outcome for the seminar and can make use of the knowledge, skills and competencies listed in the learning outcomes in the Curriculum of the Master programme.
To receive the top grade, the student must with no or only a few minor weaknesses be able to demonstrate an excellent performance displaying a high level of command of all aspects of the relevant material.
- Project work