OS4141 Advanced Managerial Data Analysis

This course demystifies AI and machine learning (ML) and prepares students to evaluate, manage, and make decisions about data-driven and AI-enabled systems in organizational and operational settings. The emphasis is on understanding how major classes of models work at a conceptual level, how data choices affect outcomes, and how managers and customers should assess performance, uncertainty, and risk. Students will learn to interpret results, diagnose limitations, and communicate implications to non-technical stakeholders for common modeling approaches. The course also introduces core machine-learning and AI terminology and concepts, with a focus on advantages and limitations of major classes of models and criteria for selecting, evaluating and governance of data-driven tools for a given application.

Prerequisite

This is a second course in statistics and data analysis. Students are expected to have prior knowledge of foundational concepts at the level of OS3160 or OS3170, including: * terminology such as variable, parameter, model, inference, estimation, prediction, error, bias; * basic summary statistics and simple linear regression; * fundamentals of probability; * limitations of statistical inferences including sample selection and measurement error. Students should be able to run and make minor modifications to existing code and should be familiar with basic programming concepts such as objects, functions, conditional statements, and loops; the ability to develop programs from scratch is not required

Lecture Hours

4

Lab Hours

0

Course Learning Outcomes

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

  1. Evaluate predictive models in terms of performance, uncertainty, and suitability for a decision context.
  2. Apply concepts such as bias–variance tradeoffs, training/validation/testing splits, and performance metrics to assess model quality.
  3. Design and critique evaluation strategies for AI-enabled systems, including chatbots and decision-support tools.
  4. Distinguish among tools including statistical models, machine learning models, optimization, LLM-based systems, and humans and articulate their strengths, limitations, and data requirements.
  5. Communicate model capabilities, risks, and limitations clearly to managers, customers, and other stakeholders.
  6. Make informed managerial decisions about data access, human oversight, and deployment of AI/ML systems.