(Updated: May 05 2025 03:10)

Under Construction

Simpson’s Paradox in 3D
Simpson’s Paradox in 3D

If it were a fact, it wouldn’t be called intelligence
— Donald Rumsfeld interviewed by Stephen Colbert
Wow!
— Stephen Colbert

Course work and grades

  • Final exam: 25% Two hours during the final exam period.
  • Midterm test: 20% Thursday, May 22, 2025, 11:30 am to 12:30 pm
  • Project: 25%
    You will work on an individual project in which you solve a real problem involving real longitudinal data and prepare a report including analyses, graphical displays and a careful interpretation of your explorations and analyses. The project has five components:
    1. Choose a problem that involves an insightful analysis of longitudinal data by May 20. Post information about the problem and data as a private post on Piazza to the instructor.
    2. An interim report on you progress submitted on May 30, which you will discuss with the instructor to get feedback.
    3. A ‘.R’ script using Markdown that produces a detailed analysis and presentation of your work, including diagnostics, etc. This output can be quite detailed. Submit a draft on June 9, final version on June 12.
    4. A ‘.R’ script using Markdown that produces an attractive and readable report with your main findings prepared in a way that would be suitable for a publication. You need to include all relevant references, data sources, etc. Aim for a maximum of 30 pages. Submit draft on June 9, final version on June 12.
    5. Prepare slides for a 10-minute presentation on June 12. The slides should be prepared with R-markdown using the ioslides format or other slide format. The 10-minute limit is strict. Be aware that it takes careful preparation and rehearsing to give a good presentation in such a short time. The presentation will be followed by a 5-minute question and discussion period.
  • Assignments: 20%
    • Combination of occasional individual and weekly assignments. Most are done on Piazza.
    • Some assignments may have a higher weight than others.
    • All team members should feel responsible for helping each other to prepare and understand all solutions.
    • For team assignments, you split the questions evenly among team members and decide which team member will be responsible to take the lead for which question(s). This is best done by agreeing to cycle through the questions in a systematic way. I will provide random numbers to help.
    • Team assignments are done in three steps. For an assignment given on Tuesday:
      • Step 1: to be completed by deadline #1, usually the following Friday at noon:
        • The team member reponsible for a question posts a tentative solution on Piazza before deadline #1.
        • It must have a title of the form specified for the assignment.
        • The solution must start by repeating the question so someone looking at the solution can tell what question it solves.
        • For math, use the LateX editor in Piazza. You can also make sketches on paper, photograph them and upload the photograph to Piazza. Use markdown in R as much as possible.
        • When you first submit the post, make it private to your team and use the folder assn X, where X is the number of the assignment.
        • Each post remains private to your team until after deadline #3.
        • You get full marks for effort in making an honest attempt, it does not have to be completely correct.
      • Step 2: to be completed by all teammates by deadline #2, usually the following Sunday at noon:
        • Provide feedback on the solutions posted by your teammates: suggestions for improvements, improving coding in R, pointing out inconsistencies or errors, broadening the answer to cover a broader range of cases, etc.
      • Step 3: to be completed by deadline #3, usually the following Monday at noon:
        • The team member responsible for a question reviews the suggestions made by teammates and incorporates them into the answer.After deadline #3 and before the next class, make the solutions public to the class.
      • I will select some solutions as interesting sample solutions and add them to the star folder. Being added to the star folder does not necessarily imply that a solution is correct, nor does it mean that it’s the best solution. It just means that I found some aspect of it interesting and illustrative of the issues presented in the question. Conversely, not getting a star does not mean that you don’t have an excellent solution. Sometimes solutions with ‘errors’ are more interesting from a pedagogical point of view.
  • Class and Piazza contributions: 5% (possibility of bonus marks for outliers)
    • Contribute actively in class and post questions and answers about course material on Piazza.
    • Contribute to the course ‘wiki’: post links to something on the web that is interesting and relevant to statistics and add a link to it on the wiki page with a brief summary of the content and relevance. You may import material from other course wikis if it is interesting and relevant to this course.
    • Add links to your contributions in your LOG file.
  • Weekly feedback every Friday evening and quiz questions: 5%
    Every Friday create a post that is private to the Instructors (it may be made public later during the weekend or you can make it public yourself) with information on each of the following:
    • What was the most interesting or challenging idea during the week?
    • What questions are you left with?
    • A quiz question based on the material of the week.
    • Be sure to choose the ‘Feedback’ folder when you post your feedback.
  • Class attendance: Since lecture videos are posted online and much of the team collaboration can take place through Piazza, it may be tempting to give class attendance a lower priority than it merits. Attendance and active participation in class improves your engagement and experience of the course, as well as that of your colleagues. There are strong positive externalities in regular attendance. I hope that all of you plan to attend at least 80% of classes. Please let me know if you expect your circumstances to prevent you from doing this.

Prerequisites

Admission to a graduate program in Statistics or appropriate permissions. It is assumed that you have basic mastery of linear algebra, vector calculus, regression and mathematical statistics, including likelihood theory.

Textbooks

  • There is no single textbook. We will use notes, slides and many references to textbooks and articles.

Getting Help

  • Post questions and comments about the course material on Piazza. Post your questions to the entire class so everyone can benefit from the discussion and answer. I will monitor Piazza and participate if other students don’t have an answer.
  • If you have a personal question for the instructor, you can post it on Piazza as a private posting to the instructor. This should only be used for personal questions that are of no interest to the rest of the class.
  • If you happen to post a private question whose answer is of general interest to the class and that contains no personal information, I will assume that you consent to it being posted to the whole class unless you explicitly request otherwise.
  • You can ask your teammates or other classmates directly.
  • You can see the instructor during office hours or after class.

Course policies

Missed deadlines

Late activities or projects may be penalized 20% of the value of the activity for each day (or portion of a day) they are late.

Academic honesty

Familiarize yourself with the York University Senate Policy on Academic Honesty. Violations of academic honesty are treated very seriously in university. Always cite your sources for any information you use. This can as simple as providing links to websites you have visited to get information.

If you use AI to help with coding or substance of an assignment or project, describe how you used it and how you verified the results.

You do not need to describe your use of AI if you have used it only for searching.