Announcements

New class time: Wed, 9:30 - 10:45, Gardner 04.

Right now (by Aug 22) everybody should e-mail me: paper title, abstract, how far along is the paper?, who are your advisors / who have you been talking to?

Check this web site (www.lhendricks.org) regularly for updates.

Syllabus

Purpose:

Dissertation stage PhD students present their work and receive feedback. Objectives:

1. Learn how to improve your paper.
2. Practice presentation skills - very important for the job market.
3. Generate ideas for additional research / new papers.

Organization:

One class meeting per week in fall and spring. Each student presents one of his/her dissertation papers twice, typically once in the fall and once in the spring. The audience provides feedback. After each session, the instructor provides written comments / suggestions. At the time of the second presentation, each student should address his/her progress towards addressing previously open issues and comments received during the first presentation. In exceptional cases, a student may present two different papers.

I assign presentation dates but you may trade them with other students. Let me know if you do so.

Please e-mail me a copy of your paper at least 1 week prior to your presentation. I will post the paper on the course web site. All students are expected to have read and thought about the paper before coming to class.

If your paper contains a non-trivial model, you should also submit a model summary which describes your model in standard format:

1. Describe demographics, endowments, preferences, technologies.
2. State the planner's problem (optional, but often useful).
3. Describe market arrangements.
4. State each agent's problem: states, choices, preferences, constraints.
5. Define a solution to each agent's problem. (A set of objects that satsify a set of equations.)
6. Define an equilibrium.

The purpose of the model summary is to (i) state clearly what the model elements are and (ii) to collect all equations in an organized manner. This avoids a lot of confusion. Note: A well-written paper should contain all of this in exactly this order.

Students are required to attend all class meetings. Classes meet Tuesdays, 11am to 12:15pm. Gardner 209.

This is the first time the class is taught. Suggestions on how to improve its organization are very welcome.

Grading based on:

1. Does the presentation reflect research progress commensurate with the student's stage in the program?
2. Does the second presentation show progress towards addressing the open issues / comments received in the first presentation?
3. Class participation.
4. Students are required to attend the macro seminars.

Schedule

Rules and Tips for Presentations

Presenters have 75 minutes sharp. Discussion may go on after the presentation. The presenter should arrive early to set up the equipment.

The audience may (should) interrupt any time to ask clarifying questions / make suggestions.

I expect everybody to behave professionally: arrive on time, be prepared, have read and thought about the paper.

Most seminar presentations are lousy, even those given by experienced speakers. Therefore, read the following recommendations.

Recommendations:

1. Focus! Cover your main points. Don't digress.
2. Short introduction. The "Minnesota rule" holds that the first model equation should be on the board within 5 minutes.
3. Forget the literature. You should have a slide, just in case somebody asks. At the same time, be very clear about what is new in your paper.
4. Summarize at the beginning:
   1. What is your question?
   2. How do you approach it?
   3. Why is this approach reasonable?
   4. What is your main finding?
   5. How does your paper change the way I think about the broad issue studied?
5. At the end of the intro, people should have a clear punchline in their minds. It has to be crystal clear what the audience should take away from the talk.
6. Answering questions:
   1. Don't guess. If you don't know, say so. Audiences lose respect for speakers who speculate.
   2. Don't get side-tracked. If a question is outside the scope of your paper, defer it for discussion after the talk.
   3. Crowd control. Some seminars get out of hand. People get hung up on a particular issue. Or they stubbornly try to convince you that your paper is fundamentally flawed. You need to exercise authority and move on. Easier said than done!
7. If you are presenting work in progress, clearly state what you have not yet done. In the class setting, I also recommend to invite suggestions for problems you are stuck with.

Common mistakes:

1. Too many slides. This is the most common mistake. For a 75 minute talk, you should expect to cover at most 25 slides.
2. Too much on a slide. The "Kehoe test" is useful: If you drop the slide on the floor, can you clearly read what is on it? If not, you have too much on the slide / your fonts are too small. (This test evidently pre-dates laptop projectors.)
3. Too long motivation. For many questions, it is obvious why we should care about it. If you need 20 minutes to explain why we should care, perhaps it's not a good project (or the wrong audience for the paper).

Also take a look at tips for writing. Each paper should contain a model description in standard format.