School of Accounting and Finance

AFM 346: Applications of Predictive Analytics in Accounting and Finance Course Outline

Course Instructor:

Name:

Office Location: Email:

Office Hours:

David Modjeska

Online [email protected] Thursdays, 4:00 p.m. – 6:00 p.m., by appointment, email

           Teaching Assistants:

  Name: Email:

Name: Email:

Celinda Ma [email protected]

Karina Qiu [email protected]

          Course Description:

Models help us to think better about the world: they provide insights into complex, large, or dynamic data. Sometimes models also predict the value of important variables - such as price, sales, and risk. "Prediction" here implies forecasting values that have not yet been categorized or that can only be measured in the future. For example, is this transaction fraudulent? As another example, how many products will be sold next quarter?

This course introduces machine learning and its applications for prediction in accounting and finance. We will cover a variety of supervised learning methods from linear regression through neural networks. A main theme of this course is building machine learning workflows - including data transformation, model training, prediction, and evaluation. We will also discuss methods for exploratory data analysis, feature engineering, and assessing results (including performance, potential overfitting, and bias vs. variance). Some application areas include loan defaults, product pricing, product sales, and market trend.

AFM 244 (or the equivalent) is a prerequisite for this course. Course assignments and projects will be completed in R.

AFM 346 —Predictive Analytics 1

Lecture Schedule:

Section Days of the week Time Instructor

001 Monday and Wednesday Monday: 7:00 p.m. – 8:50 p.m. David Modjeska Wednesday: 7:00 p.m. – 7:50 p.m.

The course will be taught remotely, including lectures and tutorials.

We will be using Piazza for online discussion. This tool is available for you to interact with and get assistance from classmates, TAs, and the instructor. Rather than emailing questions to the instructor, we encourage you to post them on Piazza. A link can be found in the Course Information module on the LEARN platform.

BAFM Program Level Learning Outcomes

Course Learning Outcomes

By the end of this course, you will be able to:

1. Define predictive analytics and machine learning. Explain the model building process, over and underfitting, bias and variance, discuss the use of independent training, testing and validation data sets, as well as cross-validation. (Fluency in the languages of business, entrepreneurship and technology)

2. Construct models and evaluate them using different performance metrics. Derive an optimal set of hyperparameters through experimentation. (Problem-solving capabilities)

3. Analyze and interpret key model performance metrics. Establish relationships between predictor and predicted variables. (Problem-solving capabilities)

4. Produce replicable reports that combine code, results, and free form text. Communicate complex relationships using tables and graphs. Write and debug code for training and evaluating models, as well as making predictions. (Communication capabilities; problem-solving capabilities)

5. Analyze, discuss, and summarize critical issues related to data analytics and machine learning, including how predictive analytics impacts businesses, ethical implications of predictive analytics, and the relevance of model explainability, among others. (Attributes/qualities of a financial professional)

                Each of the School of Accounting and Finance’s Program Level learning outcomes identifies a knowledge, skill or value of a financial professional. These outcomes are organized into seven areas as reflected in the graphic. The puzzle pieces reflect the integration of all areas. All outcomes are developed through experiential learning.

 AFM 346 —Predictive Analytics 2

Course Resources:

Readings:

Required (specific chapters only)

• Boehmke, Bradley and Brandon Greenwell. 2020. Hands-One Machine Learning with R.

• Cimentada, Jorge. 2020. Machine Learning for Social Scientists.

• Healy, Kieran. 2018. Data Visualization: A practical introduction. Princeton University Press.

• Ismay, Chester and Albert Y. Kim. 2021. Statistical Inference via Data Science. CRC Press.

• Kuhn, Max and Julia Silge. 2020. Tidy Modeling with R.

• Kuhn, Max and Kjell Johnson.2019. Feature Engineering and Selection: A Practical Approach for

Predictive Models.

• Wickham, Hadley and Garret Grolemund. R for Data Science. O'Reilly, 2017.

Recommended

• James, G., D. Witten, T. Hastie, and R. Tibshirani. An Introduction to Statistical Learning with Applications in R. Springer, 2017.

• Silge, Julia. Supervised Machine Learning Case Studies in R website.

• Tidymodels: “Learn” section of tidymodels.org website.

           Other Materials:

1. Software Licenses

1. R (pricing is free)

2. RStudio Desktop – Open Source Edition (pricing is free)

Course Evaluation:

Assessment Method

Quizzes Assignments (5) Midterm Project Final Project Participation

Date

Weekly (Weeks 1-11) Approximately Biweekly Weeks 6-7 Weeks 11-12 Throughout term

Percent of Total Grade

20% 40% 10% 20% 10%

100%

                     Total

   AFM 346 —Predictive Analytics

3

Quizzes

The quizzes will cover all material assigned in the course. The quiz questions may be either multiple choice or true/ false questions. The best 10 quizzes out of the total of 11 will be used to determine the final quiz mark.

Assignments

For homework assignments, you are welcome to discuss the problems with your classmates. However, the detailed solutions and the actual submissions must be the exclusive work of each individual student.

Projects

All projects must be the exclusive work of each individual student. Discussing the problems with your classmates is not permitted for course projects.

ADDITIONAL COURSE POLICIES:

Submission Times

Please be aware that the University of Waterloo is located in the Eastern Time Zone (GMT or UTC-5 during standard time and UTC-4 during daylight saving time) and, as such, the time for your activities and/or assignments are due is based on this zone. If you are outside of the Eastern Time Zone and require assistance converting your time, please try the Ontario, Canada Time Converter.

Late Submission Policy

All quizzes, assignments, and projects are expected to be submitted on time. The LEARN platform will accept submissions during the period posted for each deliverable – one week for each quiz, two weeks for each assignment, and two weeks for each project. Exceptions may be permitted only by arrangement with the instructor at least three days (72 hours) before a submission deadline, or in case of documented, extenuating circumstances (such as illness, bereavement, and so on).

Participation Policy:

Oral communication is an important part of learning. We encourage all students to participate actively in in-class and online discussions. Asking questions during Q&A sessions, contributing to group discussions, and answering questions in online discussions are all example of useful participation. Evaluation will be qualitative, rather than quantitative, so we encourage you to be thoughtful and professional in your comments.

Turnitin Policy:

Turnitin.com and alternatives: Text matching software (Turnitin®) may be used to screen assignments in this course. Turnitin® is used to verify that all materials and sources in assignments are documented. Students' submissions are stored on a U.S. server, therefore students must be given an alternative (e.g., scaffolded assignment or annotated bibliography), if they are concerned about their privacy and/or security. Students will be given due notice, in the first week of the term and/or at the time assignment details are provided, about arrangements and alternatives for the use of Turnitin in this course.

   It is the responsibility of the student to notify the instructor if they, in the first week of term or at the time assignment details are provided, wish to submit the alternate assignment.

AFM 346 —Predictive Analytics 4

UNIVERSITY OF WATERLOO AND SCHOOL OF ACCOUNTING & FINANCE POLICIES:

Details regarding School of Accounting and Finance (SAF) policies and University of Waterloo policies can be found on the SAF LEARN site “My SAF Community” at: My SAF Community Policy document- accessible for Learn - updated April 2022 - My SAF Community (uwaterloo.ca) within the Learn – SAF Course Syllabus – Policies for Students folder.

These policies are an integral part of this course syllabus. They have been posted on the SAF LEARN site as they are not course specific but are common for all SAF program courses. Please ensure that each term you are informed regarding these policies. They include:

School of Accounting and Finance Policies:

• Accommodations for missed assessments

• SAF Process for Requesting Accommodation for Missed Assessments

• Recording of Lectures

• Textbooks and Intellectual Property Rights

• Attendance at the Registered Section

.

University of Waterloo Policies:

• Academic Integrity

• Grievance

• Discipline

• Appeals

• Academic Offenses and Implications

• Accommodation for Students with Disabilities

• I-clickers

• Mental Health Support

• Territorial Acknowledgement

• Chosen/Preferred First Name

    AFM 346 —Predictive Analytics 5