[Logistics and Policies] [Project Guide] [Reviewing Guide]

Students in the class will conduct applied online market design projects. Students are strongly recommended to work in pairs. Projects will typically have two parts, with the first part containing more explicit instructions and the second part being more open ended and where students are encouraged to be creative. Many of these projects will require implementation and simulation of market mechanisms. The main deliverable for these projects is a project report which describes in detail what you did in the project and what you discovered. Project reports should be understandable to someone who knows the material of the class, but did not necessarily do or have the context of the project. I.e., imagine you are writing the report for a student at another university that is interested in online markets and has followed lectures, but did not attempt the project themselves.

Project Execution

Many of the projects will require analysis of data or simulation of online markets. For these studies you should implement your analysis or simulation in a programming language of your choice. You are strongly encouraged to use Python as for the size and scope of these projects Python will allow you to quickly focus in on the details of markets. You may use standard libraries. You may not use code from other sources or libraries that include specific algorithms you have been asked to implement. You should include all of your code in an appendix of your project report and it should be relatively easy to understand what you did from this appendix.

Project Reports

Project reports should generally contain the following five parts:

Introduction. Your introduction should motivate the study and summarize your observations. From the introduction it should be pretty clear what your results are and how you obtained those results.
Preliminaries. In your preliminaries section you should describe the basic setup of the study. If you have data, it should describe what that data is (for data provided to the class, you should note that this data set was provided and where it came from, but you should still describe what it is so that your report is complete). If you are doing a simulation, it should describe how you are doing that simulation and how you generated the data for the simulation. It should also describe the details of any mechanisms or algorithms that you are evaluating. Your hypotheses (if relevant) should be described here, along with plan that you undertook to evaluate the hypotheses.
Results. In your results setting you should present the results that you obtained. If visualizations of data or simulations are helpful, include these visualizations in your results section.
Conclusions. In this section you should give interpretations and conclusions of your results. It may be interesting to conjecture about whether results you observed may hold up in more general environments. You may also describe further questions that would be of interest to study that are motivated by your results.
Appendix. Include in your appendix any supplementary materials you used in your project. This includes your source code. You do not have to include data that was provided as part of the project. If you generated data, you should include the code that generated the data.

The length of your project report depends on the scope of the project. For short projects, the main body of the report, may be a page or two, for long reports with figures it may be 5 or so pages (If you find yourself writing reports that are longer than this, check in with the instructor or teaching assistants).

Project reports should be submitted as a single pdf file. Students are recommended (but not required) to use LaTeX. Overleaf is an outstanding cloud-based editor for collaborating on LaTeX documents. See further discussion here.

Submission

Submit your project reports electronically canvas. Project reports should be submitted as a single pdf file. Do not include your names on your submission. For group projects, you must join a Project Group before submitting.

Academic Integrity

Both students in a project group must contribute to the solution of project and the project report. Multiple groups may not work together. You are expected to implement the main algorithms in your project on your own. Any deviation from these policies that are caught will be addressed according to the university's policy on academic integrity violationsLinks to an external site.

Grading

Project reports will be evaluated according to the following criteria.

Appropriateness. The conducted project satisfies the requirements described by the project assignment.
Execution Quality. The authors execution of the project is high quality. Implementations are (likely to be) correct. Conclusions are (likely to be) correct.
Novelty. The open-ended part of the project was creative and its conclusions (even if negative) allow for new understanding of topics in the space of the project.
Writing Quality. The project report is well written. The structure roughly follows the recommended structure (with deviations that improve flow and readability). It is easy to understand what was done and what the conclusions are.

Scoring

Scoring on on the four criteria above will be conducted on a ten point scale with the following interpretations. Peer reviews should be scored according to these guidelines.

10 -- an exemplary report (should be used sparingly)
8 -- a solid report (good reports should receive this score)
6 -- report does not do everything that it should
2 -- report is mostly insufficient, though some aspects of it go in the right direction
0 -- does not even partially satisfy any evaluation criteria.

Scores between those listed above can be used as well for reports that are borderline between two scores.