Syllabus

Winter 2024 - 3 Credit Hours

Prerequisites

IMSE 317 or BENG 364 or ME 364

Meeting times

Mondays & Wednesdays 11:00 AM - 12:15 PM

Location: 1165 SSB (Social Science Building)

Instructor

Fred Feng (“Fred” is fine)

How can I contact the instructor?

Email is the preferred way to contact the instructor.

Email rules

  • imse440instructors@umich.edu This is the email address that you should use to help me keep my inbox organized.

  • You should include the course number IMSE 440 at the beginning of the subject line, followed by a brief description with a few words. For example: IMSE 440 - homework 1 doubts

  • Use your UMICH email

Note

Please note the instructors may not respond to

  • Emails that do not comply with the rules above

  • Canvas messages

  • Canvas comments under an assignment.

Office hours

All office hours by Dr. Feng are held on Zoom at https://umich.zoom.us/my/fredfeng

  • Wednesdays & Thursdays: 1:00 PM - 2:00 PM

  • Occasionally, the instructor may need to move an office hour to a different time. In this case, a Canvas announcement will be sent. You can also refer to our Course Calendar for the up-to-date office hour schedule.

  • If you are not able to attend during the above time, you can email the instructor with a request for an appointment.

Tip

All the office hours (and other course activities, home deadlines, etc.) can be found at the Course Calendar.

Zoom meeting policies

  • Students will be admitted to the meeting room one at a time, so that you can talk to the instructor individually (with the exception of a project team).

  • Please wait in the waiting room for your turn if the instructor is not immediately available. Students in the waiting room are admitted first-come-first-served.

What is this course about?

The course introduces you to develop statistical models and make inferences based on data. It introduces you to simple and multiple linear regression models, model evaluation and diagnosis, analysis of variance, and model selection. It also briefly introduces students to design of experiments. The course provides you with experience processing and analyzing data for engineering applications via in-class activities, assignments, and a project.

This course is designed to help you

  • develop, evaluate, and diagnose regression models

  • develop methods to make statistical inference from data

  • construct appropriate experimental designs for given problems

  • perform data analysis and build statistical models using standard software tools.

Textbooks

  • An Introduction to Statistical Learning: with Applications in Python, 1st edition, James, Witten, Hastie, & Tibshirani, Springer, 2023, PDF freely available at https://www.statlearning.com

Tools & technologies

You will need access to a laptop computer that

  • runs one of the following operating systems: Windows, MacOS, Unix

  • has internet access

  • has at least 2GB of hard drive space.

Lecture etiquette

  • Mute your mobile phones.

  • Do not engage in private conversations during a lecture.

  • We encourage you ask questions if you don’t understand something. Just raise your hand. Before asking a question, Please do think whether it is of general interest (e.g., the instructor might have just made a mistake) or it is really peculiar to you. If it is peculiar to you, then catch the instructor at the end of the lecture or come to the office hours.

  • If you have homework-related questions that requires more than a quick clarification, please come to the office hours or email the instructors.

Homework

There will be weekly homework for a total of eleven homework assignments.

You can find the homework schedule on the course Google Calendar.

Homework policies

  • All homework will be posted and submitted on Canvas.

  • Late homework will not be accepted without valid documented excuse. Barring extraordinary circumstances, the documented excuse must be available to the instructor at least two (2) academic calendar days prior to the deadline. Valid excuses include your own illness or injury, family emergencies, certain University-approved curricular and extra-curricular activities, and religious holidays.

  • It is recommended to budget enough time for submission. You can submit unlimited number of times for an assignment. Only the latest submission will be graded.

  • The excuses that will not be accepted include, but not limited to, - missed the deadline by any amount of time, - missed the deadline due to last-minute technical or non-technical issues (e.g., network, computer), - submitted incomplete file(s) (e.g., missing the data files that are needed to run your code), - submitted a wrong file(s), - submitted to a wrong assignment.

  • Your one (1) lowest homework grade will be dropped when calculating the final grade. No questions asked. An un-submitted work counts as zero.

  • Homework solutions will be posted on Canvas after the deadlines. It is recommended to go through the solutions (even briefly) for your study purposes.

Homework format requirements

  • Use the provided Jupyter notebook file as the starting point, and include all your code and answers in it.

  • Show your answers to the questions clearly and explicitly. You can use the Markdown cells to type in your answers. You can also use the print function to show the answers. The bottom line is we do not have to search for the answers buried in your code or output.

  • You will submit a ZIP file called imse440-hwX-yourUniqueName.zip, where X is the number of the current homework. For example, if I were to submit for homework 1, it would be called imse440-hw1-fredfeng.zip.

  • When I extract your compressed file, the result should be a directory called imse440-hwX-yourUniqueName, containing, at a minimum, a Jupyter notebook file called imse440-hwX-yourUniqueName.ipynb and all other files (e.g., data files) needed to run your notebook on my machine.

Attention

Homework without complete Honor Code signing will not be graded and get an automatic zero.

Attention

In your submission, you need to include all necessary files (including data files, even if you didn’t make any changes to them), so that we can run your code off the bat. We will not supply any files or make any changes before running your code on our machine.

Attention

You should make sure your submitted notebook does not generate errors. Before submission, you should click (on the JupyterLab menu) Kernel -> Restart Kernel and Run All Cells… to make sure it doesn’t generate errors. Also remember to use relative paths when specifying file locations. Any parts of your homework that generate errors (e.g., due to failing to read in data) will receive zero points.

Homework general suggestions

  • Start early! If you run into software or coding issues, it may take time to solve. It is best to find these problems early so we have time to help you.

  • Back up your work! I would recommend doing your work in a cloud file storage folder such as Dropbox (which is freely available for UM students).

  • Make your code more readable

    • Do not write all your code in a single cell or several large cells. Separate different parts of your code/answers in separate cells for better readability.

    • Use the Markdown cells to make your work more readable by briefly explaining what you are doing.

    • Additionally, use comments in the code cells to explain your code.

    • Follow naming conventions (e.g., import pandas as pd).

    • Use meaningful variable names.

Homework regrading policies

  • Regrade requests for homework must be made within ten (10) academic calendar days of when the assignment is returned and must be submitted to the instructor in writing.

  • The entire problem, not just the disputed parts, may be reviewed. Thus, it is possible to receive a lower grade than before.

  • There will be no regrading nor any grade changes after the last week of class.

Project

There will be a team project in this class. See the Project guidelines for details.

Grades

Item

Percent

Homework

50%

Project

50%

The final grades may be curved as necessary.

Note

Requests for improving grades based on individual needs will never be considered.

Honor Code policies

The University of Michigan-Dearborn values academic honesty and integrity. Each student has a responsibility to understand, accept, and comply with the University’s standards of academic conduct as set forth by the Academic Code of Conduct, as well as policies established by the schools and colleges. Cheating, collusion, misconduct, fabrication, and plagiarism are considered serious offenses. Violations will not be tolerated and may result in penalties up to and including expulsion from the University.

General course rules

  • All course work (homework, project) must represent your own work.

  • Avoiding plagiarism: You are not allowed to submit, as your own, work that is not the result of your own labor and thoughts.

    • Work (your homework, project submissions) that includes materials (e.g., texts, codes, images, figures, tables, etc.) derived in any way from the efforts of another person, by direct quotation, paraphrasing, or editing, should be fully and properly documented.

    • To avoid plagiarism, you should cite all sources of both ideas and direct quotations, including those found on the internet.

    • The citation should provide enough information so that the original source of the material can be located.

  • You are not allowed to share/use any of the lecture notes, codes, or other course materials with/from another student. If you was not able to attend a lecture or need help with lecture materials, please contact the instructor.

  • If you have any questions about whether something is or is not allowed, ask the instructors beforehand.

Homework rules

  • All the General course rules.

  • You are allowed to consult with another student from the current class during the conceptualization of a problem. However, all written work, whether in scrap or final form, are to be generated by you working alone.

  • You are required to disclose in each homework submission any person that you have discussed the homework with.

  • You are not allowed to discuss an homework with anyone outside the current class.

  • You are not allowed to possess, look at, use, or in any way derive advantage from another student’s work.

  • You are not allowed to compare your solutions, whether in scrap paper or final form, to another student.

  • You are not allowed to use the solutions or assignments prepared in prior semesters, whether the solutions were made available by the instructors in a previous semester or from a former student’ work. Note this applies to the students who have taken this class before.

  • You are not allowed to self-plagiarize (i.e., using your own previous work (e.g., homework or solutions), especially if you have taken this class before.)

  • It is your responsibility to make reasonable efforts to make sure your work is not shared with another student.

  • Penalty policies

    • A single offense will result in 0 points for the involved homework and a reduction of three letter grade levels in the final course grade (e.g., from A- to B-).

    • Multiple offenses in homework will result in a failing grade (E) for the course.

    • If a student provided unauthorized help to other students, all parties will receive the same penalty.

Project rules

  • All the General course rules

  • Your team is not allowed to receive any help on the project from another person other than the course instructors.

  • The essence of all work that you submit for your project must be your own. You are allowed to use code snippets (defined as no more than a few lines of re-usable code) from the internet or elsewhere. However, the snippet must not constitute the core part of your work. And you need to properly cite the code snippets that you borrowed elsewhere.

  • You are not allowed to self-plagiarize, meaning that you can not submit your own previously work (e.g., from another course that you took or are currently taking).

  • Penalty policies: An offense in an project may result in up to receiving a failing grade (E) for the course for all team members at the instructor’s discretion.

Warning

All Honor Code violations will be penalized to the full extent specified by the “penalty policies” section and reported to the University’s Academic Integrity Board (AIB) with no exceptions.

Student Food Pantry

The pantry exists to support individuals on their journey as they work toward achieving their goals. We are committed to increasing access to food as a key to success, by assisting any student in need! If you need access or have questions, please contact the Office of Student Life by phone at 313-593-5390, by email at umdearbornpantry@umich.edu

University-wide policies & information

Please see the Course Policies menu on Canvas for information on the following

  • University Attendance Policy

  • Academic Integrity Policy

  • Counseling

  • Disabilities Services

  • Safety Statement

  • Harassment, Sexual Violence, Bias, and Discrimination

Miscellaneous

The instructor reserves the rights to make any changes to this syllabus as deemed necessary.