MIE1626: Data Science Methods and Statistical Learning

Summer 2026

Instructor

Michael Guerzhoy
Michael Guerzhoy

Email: guerzhoy@cs.toronto.edu
Office: BA 2028
Office hours: by appointment

Course forum: Piazza. Students are responsible for reading all announcements posted on Piazza.

Course overview

MIE1626 introduces core techniques in data science through lectures, tutorials, and mini-projects. The course emphasizes statistical inference taught in a computational style, the modern data-science workflow (data acquisition, analysis, presentation, and actionable recommendations), and applied coding practice in R and Python. The lectures, tutorials, and mini-projects support students' work on a course project and prepare them for future research and professional work involving data.

The three pillars of the course are:

  • Statistical inference, taught in a more computational style than the standard first course.
  • The modern data-science workflow: agentic coding for implementation, alongside the cycle of acquiring data, analyzing it, presenting results, and proposing concrete action.
  • Coding practice in R and Python, primarily through small worked examples that students are also expected to read and understand.

Mini-projects

The course includes four mini-projects, each worth 5% of the final grade. These are coding assignments. Students may use AI tools in completing them, but they are responsible for the underlying material, which is examined in the written tests. The mini-projects are intended to be modest in scope; their primary purpose is to provide concrete content for examination.

  • Mini-project 1 (md). Posted May 16. Topics: tidyverse, probability, maximum likelihood, and Bayesian inference.
  • Mini-project 2 (md). Due May 28. Predicting ICU adverse outcomes using logistic regression, ROC analysis, and the threshold-selection problem in the “doctor with 25 charts” scenario. (See the walkthrough.)
  • Mini-project 3 (md). Due July 25. COMPAS recidivism risk scores: fairness across groups, logistic regression, a neural network and what its hidden units are doing, and a decision tree.
  • Mini-project 4. Due July 25. Redo the midterm.

Course project

Each student, or pair of students, will complete a substantial data-science project on a dataset of their own choosing. The course provides light scaffolding; students choose the topic. The project is structured around three deliverables: a preliminary proposal (May 19, 2%), a preliminary presentation and document (June 15, 8%), and the final project (August 10, 30%).

See the course project page for the full description, component requirements, deadlines, and a worked example.

Tests

There are two written tests. Both are open handout: the test reference sheet gives R syntax for the verbs used in the course and is distributed in advance, so nothing needs to be memorized. You are still expected to know what each function does and when to use it. The tests are where the material behind the mini-projects is examined.

  • Test 1 (midterm). Friday, May 29, 1 hour 10 minutes, 10%. Covers everything through Week 3. The test paper (PDF). Mini-project 4 is to redo this test.
  • Test 2 (final). Wednesday, August 12 at 2:00pm, 3 hours, 30%. Cumulative, with the emphasis on material after the midterm. The test paper (PDF).

The test coverage page lists, topic by topic, what is and is not examinable.

Logistics

  • Lectures and tutorials take place Wednesdays and Fridays, 2:00pm–5:00pm. Students are expected to attend both lecture and tutorial blocks.
  • Test 1: Friday, May 29. Test 2: Wednesday, August 12 at 2:00pm.
  • Course project due: August 10.
  • Tests are open handout. See the tests section for the reference sheet, the coverage page, and the midterm paper.
  • The course has been extended to a Y-course by request, so the final test and the project due date fall in August.
  • All course handouts are posted on this website.

Grading and schedule

The grading scheme and schedule are as follows.

ItemWorthDue
Mini-Project 15%May 16
Preliminary Project Proposal2%May 19
Mini-Project 25%May 28
Test 110%May 29
Mini-Project 35%July 25
Mini-Project 45%July 25
Preliminary Project Presentation & Document8%June 15
Test 230%Wed Aug 12, 2:00pm
Course Project30%Aug 10

Late work. Each student starts the term with 7 “grace days” that can be used to submit late work without penalty (e.g., a submission 25 hours after the deadline uses two grace days). No more than 3 grace days may be used on any single submission. Grace days do not apply to the preliminary project presentation, the tests, or the final project.

References

There is no required textbook. Lecture slides and notes are posted on this website. The following texts are recommended as references.

  • Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani. An Introduction to Statistical Learning with Applications in R (ISLR), 2nd ed. Springer, 2021. Free PDF at statlearning.com.
  • Andrew Gelman and Jennifer Hill. Data Analysis using Regression and Multilevel/Hierarchical Models. Cambridge University Press, 2006.
  • Cosma Rohilla Shalizi. Advanced Data Analysis from an Elementary Point of View. Cambridge University Press (forthcoming). Preprint at stat.cmu.edu/~cshalizi/ADAfaEPoV/.
  • Fred Ramsey and Daniel Schafer. The Statistical Sleuth: A Course in Methods of Data Analysis, 3rd ed. Brooks/Cole, 2013.
  • Kieran Healy. Data Visualization: A Practical Introduction. Princeton University Press, 2026. Free online at socviz.co.

Resources