SS 3860B Generalized Linear Models / SS 9155B Statistical Modelling II Winter 2024
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Department of Statistical and Actuarial Sciences
SS 3860B Generalized Linear Models / SS 9155B Statistical Modelling II
Winter 2024 Course Syllabus
1. Course Information Course Information
Instructor |
Day/Time |
Location |
Contact |
Dr. Camila de Souza |
Tuesday: 9:30 am – 11:30 am Wednesday: 10:30 am – 11:30 am |
NCB-117 |
use OWL messages (contact “Instructor Role”) |
Prerequisites for SS 3860B: SS 3859A/B with at least 60%
Prerequisites for SS 9155B: SS9159A
Unless you have either the requisites for this course or written special permission from your Dean to enroll in it, you may be removed from this course, and it will be deleted from your record. This decision may not be appealed. You will receive no adjustment to your fees in the event that you are dropped from a course for failing to have the necessary prerequisites.
2. Instructor Information
Instructors |
|
Office |
Phone |
Office Hours |
Dr. Camila de Souza |
use OWL messages (contact “Instructor Role”) |
WSC 225 |
519-661-2111 x83618 |
TBD |
Warning |
Students must use OWL messages to contact Dr. Camila de Souza. Messages sent to the instructor’s UWO email will NOT be replied. |
You can expect a response to a message to the instructor within approximately 48 hours during the work week (during busy times, it may take a little longer). Note that messages will not be answered within the 24-hour period before exams or project deadlines, nor can I guarantee responses over weekends/holidays.
3. Course Syllabus, Schedule, Delivery Mode
Course description
In this course, we will use the R statistical software to study both applied and theoretical aspects of different extensions to the linear regression model framework. Course topics include logistic regression, Poisson log-linear models, contingency tables, multinomial regression, mixed effect models, and nonparametric regression.
Course Objectives
By the end of this course, you should be able to:
. Select an appropriate statistical method for analyzing data with a continuous, count, binomial, or multinomial response variable.
. Explain maximum likelihood inference for the generalized linear model framework.
. Conduct different analyses in R, including computing parameter estimates and confidence intervals, conducting hypothesis tests, selecting variables, comparing competing models, and assessing goodness offit.
. Summarize and report your results for statistical and general audiences.
Tentative Course Schedule - Textbook: Extending the Linear Model with R, 2nd Edition
Week |
Topics |
Reminders |
|
1 |
Jan 8- 12 |
- Review of Linear Regression (Ch. 1) - Review of the Principle of Maximum Likelihood |
|
2 |
Jan 15- 19 |
- Logistic Regression (Ch. 2) - Binomial Regression (Ch. 3) |
|
3 |
Jan 22-26 |
- Poisson Regression (Ch. 5) |
Jan 26 Assignment 1 |
4 |
Jan 29- Feb 2 |
- Poisson Regression (Ch. 5) Modelling data from two-way contingency tables (Ch. 6) |
|
5 |
Feb 5-9 |
- Three-way contingency tables (Ch. 6) |
|
6 |
Feb 12- 16 |
- Multinomial logit model (Ch. 7) |
Feb 16 Assignment 2 |
7 |
Feb 19-23 |
Reading week |
|
8 |
Feb 26- Mar 1 |
- Generalized linear models (Ch. 8) |
Midterm exam |
9 |
Mar 4-8 |
- Generalized linear models (Ch. 8) |
|
10 |
Mar 11- 15 |
- Random effects (Ch. 10) |
|
11 |
Mar 18-22 |
- Random effects (Ch. 10) - Longitudinal data (Ch. 11) |
Mar 22 Assignment 3 |
12 |
Mar 25-29 |
- Repeated measures (Ch. 11) - Nonparametric Regression (Ch. 14) |
|
13 |
Apr 1-5 |
- Nonparametric Regression (Ch. 14) |
|
14 |
Apr 8 |
- Last day of classes |
Apr 8 Assignment 4 |
Key Sessional Dates
Classes begin: January 8, 2024
Winter Reading Week: February 17 – 25, 2024
Classes end: April 8, 2024
Exam period: April 11 – 30, 2024
Contingency plan
Although the intent is for this course to be delivered in person, should any university-declared emergency require some or all of the course to be delivered online, either synchronously or asynchronously, the course will adapt accordingly. The grading scheme will not change. Any assessments affected will be conducted online as determined by the course instructor.
4. Course Materials
Required text: Faraway, J. J. (2016) Extending the Linear Model with R: Generalized Linear, Mixed Effects and Nonparametric Regression Models, 2nd Edition. CRC Press.
https://julianfaraway.github.io/faraway/ELM/
Additional (non-required) text: Roback, P., & Legler, J. (2021). Beyond multiple linear regression: applied generalized linear models and multilevel models in R. CRC Press.
https://bookdown.org/roback/bookdown-BeyondMLR/
R statistical software package: This course is heavily based on R; therefore, all assignments will
require coding in R and Rmarkdown. Please ensure the latest R version is installed on our computer and R studio (https://posit.co/download/rstudio-desktop/).
OWL
Students are responsible for checking the course OWL site (http://owl.uwo.ca) on a regular basis for news and updates. This is the primary method by which information will be disseminated to all students in the class. If students need assistance with the course OWL site, they can seek support on the OWL Help page. Alternatively, they can contact the Western Technology Services Helpdesk by phone at 519- 661-3800 or ext. 83800.
5. Methods of Evaluation
Component |
Weight |
Deadlines/Due dates |
Assignment 1 |
5% |
January 26 |
Assignment 2 |
5% |
February 16 |
Assignment 3 |
5% |
March 22 |
Assignment 4 |
5% |
April 8 |
Midterm (2 hrs) |
35% |
to be scheduled by the Registrar’s Office Tentatively scheduled for February 27 |
Final exam (cumulative, 3 hrs) |
45% |
to be scheduled by the Registrar’s Office |
Assignments
- Assignments will be available on the course OWL site. However, you will not submit your solutions to OWL. Instead, assignments must be submitted through Gradescope, an online collaborative grading system – there will be a link to Gradescope on our OWL course site. You are responsible for ensuring that your assignment is successfully uploaded and legible.
Submissions that cannot be read by the grader will receive a grade of zero. Assignment questions must be properly assigned to each page, or the submission will not be graded.
- After receiving the grades from an assignment, students will have seven days to submit any
regrade requests on that assignment, with the exception of the final assignment. After this seven-day period, regrade requests will NOT be accepted. Regrade requests must be made using the Gradescope tool “Regrade Request” .
- Assignment submissions are due 11:55 pm (Eastern Time) on the due date. Assignments that are up to 24 hours late will receive a deduction of 15% on their mark unless late coupons are used. Late assignments up to 48 hours will receive a deduction of 30% on their mark unless late coupons are used. No extensions will be given beyond 48 hours.
- Each student will have 2 late coupons worth 24 hours each that they can use at their own
discretion for whatever reason (Note: I do not need to know the reason) towards their assignments. You can use them together on one assignment for a 48-hour extension or on two separate assignments for a 24-hour extension each. No extensions will be given for any reason beyond the use of the late coupons; therefore, I suggest saving them until you absolutely need them. You do not need to tell me when you will use the coupons; they will automatically come off the first two late days.
- Solutions to assignments will not be posted; however, TAs will provide comments on incorrect answers using Gradescope, which will allow students to find out the correct solutions. In addition, students can ask the instructor and TAs for more details on solutions via the Regrade Request tool on Gradescope and during office hours.
Midterm and final exams
- There will be a 2-hour in-person closed-book midterm exam. It is tentatively scheduled for February 27, 9:30 am to 11:30 am, to be confirmed by the Registrar.
- There will be a 3-hour in-person closed-book final exam, and its time will be scheduled by the Registrar’s Office.
Rounding of marks
Across the Sciences Undergraduate Education programs, we strive to maintain high standards that reflect the effort that both students and faculty put into the teaching and learning experience during this course. All students will be treated equally and evaluated based only on their actual achievement. Final grades on this course, irrespective of the number of decimal places used in marking individual assignments and tests, will be calculated to one decimal place and rounded to the nearest integer, e.g., 74.4 becomes 74, and 74.5 becomes 75. Marks WILL NOT be bumped to the next grade or GPA, e.g. a 79 will NOT be bumped up to an 80, an 84 WILL NOT be bumped up to an 85, etc. The mark attained is the mark you achieved and the mark assigned; requests for mark “bumping” will be denied.
2024-04-13