MMA Course descriptions

Course co-ordinator: Mikhail (Mike) Simutin, Associate Professor of Finance

This course will introduce the students to the key functional areas of management and the typical decisions they face.  The course will illustrate how each functional area approaches some common managerial problems, and where data and analytics may be usefully employed.  The course will provide a framework for both the analytical tools and specific managerial problems discussed in subsequent courses in the MMA program.

Objectives of the course:

• Provide students with a general overview of the key functional areas of management and the main decisions they face
• Provide an overview of key concepts and terms in each functional area
• Provide examples of how different functional areas approach typical managerial decisions
• Provide an overview of various uses of analytics in managerial decision problems. Students should understand how “hard” analytical approaches can be combined with “softer” decision analysis to arrive at effective decision recommendations.
• The course will allow a student to construct a functional area—a managerial decision “map” where various analytical approaches may be slotted. This map will serve as a reference point for the subsequent courses.

Instructor: David Soberman, Professor of Marketing

The goal of this course is to introduce the students to key ideas about data-intensive business decision-making. Key ideas explored in the course include:

• The difference between what the data “say” and what the data “mean”
• Understanding and measuring randomness and its implications; different sources of randomness (inherently random outcomes vs. measurement errors)
• The importance of mapping out the data generation process
• The numerous ways to obtain or collect data
• Understanding various biases in data and their implications on analysis
• The value of experiments
• Differences between various modeling types

Objectives of the course:

The course is built upon basic probabilistic concepts already familiar to students (e.g. distributions, measures of variability and co-variability, standard errors and statistical hypothesis) and provides them with techniques to apply these concepts in order to facilitate robust, data-driven decision-making.

Coordinator: Opher Baron, Distinguished Professor of Operations Management

The course will be composed of short  modules (“colloquia”) taught by practitioners in the related fields. Each module will be delivered in 2-3 sessions of 2-3hrs each and will include at least one graded assignment. The course will provide students with skills that will be instrumental to achieving career success in data science and management analytics. The course will start in the fall term of the MMA program and continue through the winter term.

The colloquia that are planned for expected to be offered in 2019/20 include (please note that the list below is subject to change):

1. Ethics in Data Analytics and AI
2. APIs and Google Analytics
3. Hadoop for Data Science
4. Social Network Representation
5. Probabilistic & Bayesian Networks
6. Analytics for Customer Relationship Management
7. Analytics for Detection of Fraud and Money Laundering

Objectives of the course:

The goal of this course will be to expose students to current topics and themes in data science and management analytics.

Instructor: Opher Baron, Distinguished Professor of Operations Management and Gerhard Trippen, Associate Professor of Operations Management and Statistics

In this practicum course, you will learn how to apply model- and data-based decision making to a problem that a real organization currently faces. These problems are not only more realistic than the problems you will face in individual courses, they are more holistic. Rather than focusing on an individual component of an analytical task, they involve all key steps in a typical management analytics project, from understanding the underlying managerial issues, to proposing an effective analytical solution, constructing a modeling plan, identifying the required data sources, structuring an analytical data view, executing your modeling and data plans, and, finally, presenting your findings and proposed implementation plans.

Students will be broken up into teams of 3-4 and assigned to one of projects proposed by host organizations, working alongside an internal analytics team, while periodically meeting with their faculty coach. The course starts in September and runs until April.  During the Fall term the students will primarily focus on creating a project proposal, data request and modeling plan.  During the Spring term, the students will execute on this plan and summarize their results.  There will be internal (host organization) and external (Rotman) presentations in each term.

Objectives of the course:

The objective of the practicum course is to improve students’ skills in all key steps of a management analytics project: understanding the managerial background, structuring the project, working with data, producing relevant results, presenting them effectively, while managing the project effectively along the way.

Instructor: Binyamin Oz

This course will expose the learner to a broad range of technical skills that are required to prepare data for advanced analysis. Using a combination of theory and practical exercises and case studies, the learner will develop the data acquisition and preparation skills that are a necessary pre-requisite to applying advanced statistical modelling, data mining techniques, or machine learning algorithms to their data.

Objectives of the course:

• Demonstrate the ability to prepare, explore and validate sample data for advanced analysis
• Develop and implement BI (Business Intelligence) Dashboards to support business decision-making

Instructor: Ryan Webb, Associate Professor, Marketing Area

This course provides a hands-on introduction to the wide variety of models and techniques used in predictive analytics, including linear and non-linear regression models, classification algorithms, machine-learning techniques like SVM and reinforcement learning, and causal inference. There will be an emphasis on conceptual understanding and interpretation of the models, so that students can interpret the results of these techniques to support effective decision-making. The course will be complemented by many hands-on exercises using the R programming language.

Objectives of the course:

• Expose students to the application of predictive analytics, big data, machine learning, and decision analysis techniques in a variety of business decisions
• Enable students to:
  • Structure business decisions as analytical problems
  • Identify which data sources are needed to provide an answer
  • Understand how the data should be structured for analysis
  • Use data transformation and manipulation techniques
  • Apply appropriate analytical tools
  • Obtain insights from the results and be able to apply these insights to the managerial problem at hand
  • Communicate findings effectively

Instructor: Gerhard Trippen, Associate Professor of Operations Management and Statistics

This course will introduce the students to a diverse uses of big data techniques. These techniques are often aimed at identifying and quantifying various structures in the data (e.g. What are the key similarities between certain business units with respect to customer satisfaction? What are the characteristics of important customer segments?). Model validation and effective communication of model-based results will be stressed. The course will employ a “white-box” methodology, which emphasizes an understanding of the algorithmic and statistical model structures.

Objectives of the course:

To develop the students’ ability to:

1. Clearly explain why a particular method or algorithm is needed
2. Understand how a method or algorithm works
3. Follow the logic of an algorithm or method step by step
4. Gain a white-box insight into the inner workings of the method or algorithm
5. Apply a method or algorithm to a large, real-world data set

 

Instructor: Dror Hermel, Assistant Professor, Operations Management and Statistics

The emphasis of the course will be on systematic, logical thinking, problem solving, and risk analysis, using spreadsheets as our primary tool. We will start with the basic techniques of good spreadsheet modeling and organization, and proceed to introduce a variety of modeling techniques and approaches.  These will be illustrated by building and analyzing problems in finance, marketing, and operations.  While the underlying concepts, models, and methods of this course are mathematical in nature, we will develop them on the more intuitive and user-friendly platform of spreadsheets, always focusing on the ideas and insights, rather than the underlying mathematical details.

Objectives of the course:

• To develop students’ ability to communicate effectively with spreadsheets in the area of fact-based, data-driven, quantitative decision making
• To develop and reinforce students’ probabilistic modeling skills—when limited or no data exists to estimate effects of planned business decisions, and decision support must rely on explicit probabilistic assumptions
• To develop students’ modeling skills in the areas of optimization and simulation modeling

Instructor: Brian Keng, Adjunct Professor & Executive-in-Residence, Data Science & Kanchana Padmanabhan, Director of Machine Learning Solutions at Kinaxis and External Instructor

This course will cover some of the latest advances in Artificial Intelligence and Deep Learning and how they can be used in a wide variety of marketing applications. It will introduce students to the fundamental concepts of neural networks and deep learning, provide hands-on practice with various marketing datasets, and showcase a wide range of applications from image recognition to natural language processing. These techniques will be applied to a variety of marketing applications such as recommendation engines, customer comments analysis, targeting, churn, segmentation and lifetime value.

 

Instructor: Sridhar Moorthy, Professor of Marketing

This course is about how to use data to answer marketing questions. The questions we examine are the quintessential marketing ones: How do I identify my target segment?  How do I effectively position my product?  What features should I include in my product prior to its introduction? What is the price-elasticity of demand for my product? Is my advertising effective? What is it doing? Are consumers brand-loyal? How can I measure the value of my brand?

Objectives of the course:

This course will teach students how to apply widely-used techniques in marketing analytics to business problems. Students will learn how to use these techniques, along with real-world data, to achieve important marketing objectives such as: effective construction of market segments; effective product positioning; effective product design; measurement of price elasticity and brand value; measurement of advertising effectiveness.

Instructor: Aida Wahid, Associate Professor of Accounting

This course will build on the tools, skills, and concepts developed in the first half of the program. As an applied course, students will be expected to routinely perform accounting-based empirical analysis by using the analytics skills they have learned (e.g. SAS, R, and Python). Students must practice their ability to formulate appropriate empirical research questions in the context of the business problem or opportunity. Specifically, they will first learn how to approach and appreciate accounting information and then take advantage of the rich accounting and finance dataset to help businesses solve various problems or enhance corporate profitability. At Rotman, we have an abundance of financial accounting data including COMPUSTAT, CRSP and IBES to address a large variety of business, finance, and accounting questions. The course has four modules: 1) understanding accounting information, 2) use of financial information in the equity market, 3) use of financial information in the debt market, and 4) use of disclosure. 

Objectives of the course:

At the end of the course students will: (1) better understand and appropriately use accounting and other financially-related data, (2) more confidently conduct empirical modelling to make decisions and solve the problem at hand, and (3) appreciate the strengths and limitations of empirical analysis.

Instructor: Vahid Roshanaei, Assistant Professor, Teaching Stream, Operations Management & Statistics Area

Operations and supply chain management functions are heavy analytics users in a number of industries, including retail, transportation, healthcare, and financial services. This course will focus on identifying, developing, and applying effective analytics models and tools to solve typical operations and supply chain management problems, including network design, inventory management, assortment and price optimization, and service process design.

Objectives of the course:

Knowledge: Learn key concepts, models, tools and techniques for a range of typical problem settings in operations and supply chain management.

Problem-solving skills: Develop ability to systematically identify, create, and apply effective analytics models and tools to solve typical operations and supply chain management problems. These problem-solving skills include the ability to:

• Identify appropriate data needs
• link data to models and decision tools
• select and/or build the models and tools that best fit a business situation
• identify important mismatches between a model and the real system
• generate insightful model analyses and solutions
• correctly interpret the results
• make appropriate recommendations
• effectively justify and communicate the recommendations so they are convincing for implementation

 

Instructor: Akram Khaleghei, Senior Data Scientist, Scotiabank

In this course, students will learn decision making in different service industries including education, healthcare, retail, and car sharing using analytics. We will use Python (and other tools as necessary) to explore the application of standard analytics tools in the context of the service industry.

Students will learn how to (i) design experiments effectively, (i) apply appropriate hypothesis testing to confirm statistically significant variables, (iii) handle imbalanced datasets, (iv) develop appropriate predictive models to optimize planning, (v) implement text mining and natural language processing to predict sentiment and emotion, and (vi) how to use simulations to estimate waiting time. We will discuss how analysis of these issues supports decision making and management in different services.

This content will be introduced to students with cases and data that permit them to experience the different tools. This exposure will improve students’ confidence in using analytical tools appropriately in the context of real-world problems in the service industry.