There are 60 points in total for this class. The final grades are given as follows:
Learning outcomes will be assessed based on the performances within each of the following categories:
Bonuses may be obtained at our discretaion for the groups or individuals that give outstanding work for each of these categories, described in more details below.
No final examination for this class. The learning outcomes are continuously assessed during the semester with the forecasting competitions, the reports, the presentation and the participation.
The forecasting competitions is the core of this class as it provides an opportunity for the students to combine content learned throughout the course for direct use in some realistic application. All the forecasting competitions are conducted in groups. They consist of giving forecasts of events that have not occured yet. Students are constrained to use a specific methodology covered in-class, but on top of that, they are also free to use other techniques of their own choices (it can be models seen in-class, in other classes or from their personal experience). Students are free of choice on the mean employed to produce the forecasts (we use R during the class, but students may consider alternatives). There are four competitions introduced in-class at different time during the semester (see the case studies page for more details). There are 5 points for each competition. The points are gained solely based on the performances of the group. Each competition has its own performance criterion. The group with the overall best performance receives a bonus of 1 point per competition. Be careful, the instructor has one group as well
At the end of each competitions, groups will have a week to make a short report (max 4 pages) that summarizes their experience in the competition and the different strategies they adopted. There are five points per report (four competitions). The 5 points of the grade are allocated according to the following criteria:
For the content, the reports should be structured as follows:
Code, additional tables and/or figures may appear in an appendix.
Late submission is penalized by 1 point every 24 hours after the deadline.
Failure to comply to the rules (e.g. max of 4 pages) is penalized.
Every group will give once a 10 minutes (not more!) presentation on one of their experience in the forecasting competition to the class. Every member must talk during the presentation. The goal is to share with the rest of the class your unique experience (bad or good) so every one can benefit. The 10 points of the grade are allocated according to the following criteria:
The presentations are peer evaluated: every group grades the others and send a brief report (1 page max) to the instructor. The instructor grades as well the project. The final points obtained for the project is given by taking 50% of the instructor’s grade and 50% of the average from the other groups’ grades.
In order to earn full credit for the Piazza portion, each student should make 1 or more substantive posts per week (except during the first week) related to the content of the course. A post can be a reply to another student’s question. Grading is based on Piazza meta-data that can only be accessed by an instructor.
Due to the structure of this class, attendance is very important. Moreover, students will be encouraged to work in teams on in-class activities each week, so others are counting on you to be in class and contribute. Students with valid excuse should notify the instructor as soon as possible and provide a minimum of one week notice.
Each lecture will have its own magic word that you are responsible for entering into a form link that will be provided on Piazza. In order to be eligible to receive attendance points, you must be in class, enter the magic word within 5 minutes after it is given, and stay for the entire duration of the class.