Generalized Linear Models MATH 523 (4 credits) Mc , WT 2022 Course Outline
Instructor:
Teaching Assistant: Course Website:
Johanna G. Nešlehová
Copyright By PowCoder代写 加微信 powcoder
https://www.math.mcgill.ca/neslehova/
Course Contents
A brief review of the linear model. Exponential families and link functions. Generalized linear models: maximum likelihood, iteratively weighted least-squares, quadratic algorithms for maximum likelihood, asymptotic distribution of likelihood estimators, deviance statistic, residuals. Models for count data: Poisson regression and log-linear models, goodness-of-fit, contingency table analysis, structured log-linear models, iterative methods for model fitting, overdispersion and underdispersion. Models for binomial and multinomial data: binomial and logistic regression, 2 × 2 tables and case-control designs, overdispersion, multinomial responses, conditional logistic regression, matched pairs. Special topics (if time permits): gamma regression, quasi-likelihood and generalized estimating equations, estimation of the link function.
Learning Outcomes: By the end of the course, you should have a good understanding of the theory and methodology of generalized linear models. You should also be able to rec- ognize situations where these models are appropriate, fit them to real data using statistical software and critically assess the fit.
Literature: There is no required textbook for this course. The recommended book is • Foundations of Linear and Generalized Linear Models, (1st edition)
A. Agresti, Chapters 4-7 (and parts of Chapter 8 if time permits). Other books that you may find helpful include:
• Categorical Data Analysis, (any of the three editions), A. Agresti.
• Generalized Linear Models, Chapman & Hall (2nd Ed.), P. McCullagh and J.A. Nelder. • Extending the Linear Model with R, Chapman & Hall, . Faraway.
Software: R. This is a free software which is available at http://www.r-project.org/. Please install this software during the fist week of lectures. We will use R throughout, in lectures, practice problems, assignments, and exams.
Prerequisites: MATH 423 or MATH 533 or EPIB 697. We will not be using any theory of linear models in this course as we will develop our own, but you should be familiar with regression modeling on entering this course.
Generalized Linear Models MATH 523 (4 credits) Mc , WT 2022 Course Delivery Method
Contact Activities:
• Flipped Classroom Activities: Each week, there will be two sessions (on zoom until January 24, in person thereafter if the sanitary situation allows it), from 2:35 to 3:55 PM EST on Tuesdays and Thursdays. These sessions will include one or more of the following activities:
– Summary of the pre-recorded lecture.
– Questions and discussions about the pre-recorded lecture. – Problem solving.
– Live demonstrations with R.
– R coding sessions.
You will know at the onset of each week the detailed content that week.
• Office Hours: There will be two one-hour office hours via zoom, on Tuesdays and Thursdays from 1:00 to 2:00 PM EST.
Noncontact Activities and Materials on MyCourses:
Detailed course plan for each week.
Pre-recorded lectures
PDFs of lecture notes
Video recordings of flipped class activities
Solutions to assignments, after their due date R Code
MyCourses Discussion Board
Materials for midterms and final exams
Marking Scheme: The maximum of
• 20% Assignments + 10% R Coding + 5% Class Participation + 20% Midterm + 45%
• 20% Assignments + 10% R Coding + 5% Class Participation + 65% Final
Generalized Linear Models MATH 523 (4 credits) Mc , WT 2022
Course Delivery Method
Pre-recorded Lectures: Each pre-recorded lecture lasts about 50 minutes on average, and imitates blackboard teaching, so the pace is slow. Some lectures are shorter if they will be complemented with R illustrations live. The lectures will be posted on Fridays ahead of the week in which we will discuss them, so that you can listen to the videos at your leisure and either take your own notes or add comments to mine (which will be provided along with the videos). During week X, please make sure to listen to lecture Xa before the class on Tuesday and to lecture Xb before the class on Thursday. You do NOT need to understand everything perfectly in order to attend class, it is only important that you have listened to the video. Flag whatever is unclear, and come to class to ask questions and discuss the content with me and your classmates.
Class Participation: To earn your participation grade, you need to engage in a teamwork activity. Your team will be responsible to prepare a 10 minute summary of one specific pre- recorded lecture, and prepare 2 questions about the content of this lecture for your fellow students. Details are provided on MyCourses under Week 1.
R Coding. There will be 8 R coding sheets and you will be able to complete almost all of the coding in class. Each coding sheet will be posted in MyCourses and distributed via email from Crowdmark; you will then use this link to submit your work for marking. Coding sheets for week X are due on the Monday of week X+1. No late submissions will be accepted for partial credit. To calculate the final grade, two coding sheets with the lowest grade will be dropped.
Assigments: There will be 4 Assignments during the semester which will contain easier practice problems as well as 1-2 more challenging questions so that you can practice syn- thesizing the material. You are encouraged to work together with your classmates, but you must write-up and submit your solutions individually. Copying without understanding is academic fraud. Each assignment will be posted in MyCourses and distributed via email from Crowdmark; you will then use this link to submit your work for marking. No late assignments will be accepted for partial credit.
Midterm & Final: Online and timed. The midterm will be on February 24, and posted so that you can work on it during class time. It is intended for 1h and 20 minutes. The exams will be distributed and submitted via Crowdmark. These exams are expected to be completed entirely independently.
In view of the flexible grading scheme detailed above, there will be no opportunity for a makeup midterm and no makeup work in lieu of any aspect of the course assessment. However, reasons for missing the final exam will be considered carefully and may result in a makeup exam; if granted, the latter would then be scheduled by the university according to the standard deferral process.
Generalized Linear Models MATH 523 (4 credits) Mc , WT 2022 Communication
MyCourses Discussion Board: If you have a question of mathematical or statistical nature, or a question that everyone would benefit from knowing the answer (e.g. typo in the homework, questions from lecture, questions about course administration), do NOT email me about it. I will answer all such questions on the MyCourses Discussion Board, so that everyone can see the response. There is an option to post anonymously, so long as we maintain this as a respectful space. Also, you are welcome to answer any questions of your classmates (I will be monitoring the activity), and if you participate by answering questions correctly, there can be a small boost to your final letter grade (+0.5 to the final empirical grade).
E-mail: Please use my personal email only if you need to contact me for serious personal reasons (such as a disability that requires accommodations) or if you need to schedule an appointment outside of class time or office hours. I do not have the capacity to answer a large number of emails so please make use of this resource sparingly.
Health and Wellness Resources at Mc well-being is a priority for the university. All of our health and wellness resources have been integrated into a single Student Wellness Hub, your one-stop shop for everything related to your physical and mental health. If you need to access services or get more infor- mation, visit the Virtual Hub at mcgill.ca/wellness-hub or drop by the Brown Student Services Building (downtown) or Centennial Centre (Macdonald campus). Within your fac- ulty, you can also connect with your Local Wellness Advisor (to make an appointment, visit mcgill.ca/lwa). You may also consult the Math & Stat Department EOSW Website:
https://www.mcgill.ca/mathstat/eosw.
If you have a disability and need special arrangements, please contact the Office for Students with Disabilities at 514-398-6009. If you need me to make special arrangements, please contact me as soon as possible.
Guide of Conduct for Remote Teaching
Portions of this course during which the instructor is lecturing on zoom will be recorded and made available in MyCourses to students registered in the course. However, portions of the course during which the instructor interacts with participants, e.g., question periods and virtual office hours, will not be recorded.
Generalized Linear Models MATH 523 (4 credits) Mc , WT 2022
Participants will be notified through a ‘pop-up’ box in Zoom if a lecture or portion of a class is being recorded. All participants will be muted during such recordings and to protect individual privacy, cameras should be turned off. At no point in the course will a participant be asked to turn on his/her/their camera.
By remaining in sessions that are recorded, participants agree to the recording, and they understand that their image, voice, and name may be disclosed to classmates. Participants also understand that recordings will be made available in MyCourses to students registered in the course.
The university is committed to maintaining teaching and learning spaces that are respectful and inclusive for all. To this end, offensive, violent, or harmful language arising in the context of Zoom sessions or MyCourses discussion may be cause for disciplinary action.
For additional information, please refer to the Guidelines for Instructors and Students on Remote Teaching, Learning and Assessment.
Mc Statements
• End-of-course evaluations are one of the ways that McGill works towards maintaining and improving the quality of courses and the student’s learning experience. Partici- pants will be notified by e-mail when the evaluations are available. Please note that a minimum number of responses must be received for results to be valid.
• In the event of extraordinary circumstances beyond the university’s con- trol, the course contents, delivery mode, and evaluation scheme are subject to change, provided that there be timely communications to the students regarding the change.
• Mc values academic integrity. Therefore, all students must understand the meaning and consequences of cheating, plagiarism and other academic offences un- der the Code of Student Conduct and Disciplinary Procedures. For more information, see, e.g.,
http://www.mcgill.ca/integrity/ http://www.mcgill.ca/students/srr/honest/
• In accord with Mc ’s Charter of Students’ Rights, students in this course have the right to submit in English or French any written work that is to be graded.
• Work submitted for evaluation as part of this course may be checked with text match- ing software within MyCourses.
Generalized Linear Models MATH 523 (4 credits) Mc , WT 2022
• This format for the delivery of this course is unusual. It is dictated by the current extraordinary circumstances, and aims to allow students to complete this term with the requisite knowledge for this course, and to succeed in their assessments. Everyone’s collaboration and cooperation is requested in ensuring that all videos and associated material put as the students’ disposal for the purpose of this course are not reproduced or placed in the public domain.
According to Article 18a of the Code of Student Conduct and Disciplinary Proce- dures, “It shall be an offence knowingly to procure, distribute, or receive, by any means whatsoever, any confidential academic material such as pending examina- tions or instructor-generated materials (e.g., handouts, notes, summaries, exam questions, etc.) or documents without prior and express consent of the in- structor”. Violations of this article, for example posting notes or exams online, may result in sanctions including, but not limited to, substantial marks penalties.
• AstheinstructorofthiscourseIendeavortoprovideaninclusivelearningenvironment. It may happen inadvertently that some of the course content, e.g., the topic addressed in data illustrations, is disturbing for some students. Please do not hesitate to contact me if they have specific concerns about this. The Academic Considerations Framework due to COVID-related incidental absences remains in place for the Winter 2022 term. If students are requesting accommodations, they are to be directed to fill out the form on Minerva (found under their Personal menu). The pandemic has been very difficult on all of us, so I have strived to design the course in a way that you can follow most of it remotely if the need arises. If you experience barriers to learning in this course, do not hesitate to discuss them with me and the Office for Students with Disabilities.
• Mc wishes to acknowledge the fact that it is situated on the traditional territory of the Kanien’kehà:ka, a place which has long served as a site of meeting and exchange amongst nations. McGill recognizes and respects the Kanien’kehà:ka as the traditional custodians of the lands and waters on which the university is located today.
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com