This course is designed as a follow-up to the SOM MBA core statistics class. Its goal is to give you lots of practice analyzing data and presenting the results of your analysis relying solely on the tools introduced in the core. It will revolve around weekly, open-ended "consulting assignments" of the form: "Your client is X. They give you dataset Y. They would like to know Z. Please prepare a short presentation succinctly summarizing and justifying your answer." Class time will be split among three activities: lab time devoted to discussing issues/problems that arise in the consulting assignments; mini-lectures covering issues related to data analysis and visualization, as needed; and discussion of consulting assignment presentations.
Office Hours
If you would like to meet outside of class, please email me (peter.schott@yale.edu) a few times that work for you and I will pick one. Please feel free to email questions at any time; I will try to return them promptly.
Teaching Assistant
TBA
This Course vs the SOM Core
The SOM statistics core covers hypothesis testing, ordinary least squares (OLS) regression and use of variable transformations, dummy variables and variable interactions in conjunction with OLS. This course will provide brief reviews of these powerful tools, as well as further discussion of how they can be used in many interesting ways to answer interesting policy and managerial consulting questions. Where relevant, we will discuss the relationship between these topics and more advanced data-analysis electives offered at SOM.
Software
You are to use whatever software package you like. I use STATA, so the slides will focus on that software package. The Yale StatLab has excellent tutorials for all major software packages, including R. Go to https://marx.library.yale.edu/data-gis-statlab/statlab.
Meetings
"Lab" meetings will be devoted to discussion of questions that might arise as you try to complete one of the consulting assignments, as well as mini-lectures discussing techniques that might be useful in completing the assignments. The bulk of the time in "presentation" meetings will be devoted to students' client presentations; any time left over will be used as lab time.
Groups
All assignments in this course are to be done in groups of 4. Please put yourselves into teams of 4.
Consulting Assignments
A primary goal of this course is to give you practice analyzing data, so it is expected that all team members actively participate in completing all assignments. Another goal of this course is to give you practice presenting the results of your analysis. However, since presentation time will be short, it is ok for group members to share presentation duties across cases.
NOTE
NOTE: In some instances I have used material on the web (academic papers, policy briefs, newspaper articles) to construct the assignments. To ensure the best learning experience, you are asked NOT to read any of those analyses until after you complete and hand in the assignment. Also, constructing good consulting assignments is quite difficult, so I re-use them over time. As a result, please refrain from talking to any prior participants of the course about the assignments until after you have handed them in. It is a violation of academic integrity to rely on any work from prior iterations of this or any related course in working on your assignments for this iteration of the course. Plus, it wouldn't be as much fun.
Presentations
Presentations should contain less than 10 slides (not including an optional appendix) and must be delivered in under 10 minutes. Appendix slides are to be used for backup only, e.g., in response to a request for more detail. Your slides should provide an answer to the client's question as well as a justification for why it is the best answer among alternatives. Expect to be interrupted by the client (me, your fellow students) with questions. Given that presentation time is limited, you must make sure that each slide (and each element of each slide) is necessary for your answer as well as clear. Some tips:
NOTE
Speak from the perspective of the person hearing the briefing, not from the perspective of the data
Be sure to answser the client's question
Create data displays that have a clear message, and transparent and legible axes titles and scales
When questions arise, restate the question in a way that engages everyone in the room
Though not required, feel free to gather additional data or background in conducting your analysis
NOTE: This year I will ask all students to rate all presentations on a scale of 1 (low) to 5 (high). These assessments will factor into the instructor's overall assessment for the purposes of grading.
Grading
The six assignments will be weighted equally.
Academic Integrity
No student's name should appear on a group project if the student has not contributed to the production of the project. The following is an example of unacceptable conduct: Neo agrees to produce a case write-up, putting Trinity's name on the case write-up. Trinity agrees to repay Neo by producing a subsequent assignment on his own.
NOTE
In group work, all group members are responsible for the integrity of work that is submitted. Group members should always question other members about the source of material and analysis that is being included in group projects. If a group member has any concerns about the integrity of material being submitted by the group, that group member should discuss those concerns with the instructor.
NOTE
Note that I re-use assignments over time. It is a violation of academic integrity to rely on any work from prior iterations of this or any previous course that has made use of the materials for this course in working on your assignments for this iteration of the course.
Plagiarism
The members of any academic community are expected not to present ideas or material from other sources as their own. In the context of this course, it is acceptable to, for example:
NOTE
Refer to concepts, frameworks, and analytical tools from the readings or class lectures without citation. For example, you may refer to productivity, the laffer curve, keynesian assumptions, and so on without citing the source.
However, it is not acceptable to quote or paraphrase analysis from another source and present it as your own, where "another source" includes current classmates and past students of the course. There is a difference between consulting a source and then forming your own opinion based on what you have read, and "lifting" material directly.
AI/LLM Policy
Students are encouraged to use AI/LLM's to help further their understanding of macro. For example, it can be quite instructive to ask these models to describe various macroeconomic concepts, and to generate lists of questions that can be used to test your knowlege. Please be aware, however, that these models still make mistakes from time to time. With repsect to Plagiarism, it is not acceptable to quote or paraphrase analysis from AI/LLM's as your own.
It is NOT acceptable to to quote or paraphrase analysis from AI/LLM's as your own. Any use of AI/LLMs on any assignment for this course must be noted as follows: prompts and responses should be added to presentations' appendix slides.