Data-Driven Decision Analysis

Catalog description

Business managers and public policymakers who make good decisions are in high demand and are richly rewarded. Increasingly, those decisions must be made in dynamic, data-rich environments. In those environments, having an extensive analytical toolkit and being able to build and use decision models are essential for success. Building on the foundations laid by prior coursework, we will cover advanced analytical topics from the decision sciences with an emphasis on model building. Topics may include utility function elicitation, optimal decision making under uncertainty and imperfect information, valuing flexibility with real options, portfolio theory, artificial intelligence (AI) and evolutionary computation methods, robust decision making, and Monte Carlo simulation and variance reduction methods. The focus of this course is normative, rather than descriptive decision making. The course will make extensive use of Microsoft Excel and students are expected to possess a high level of numeracy upon enrollment. Although we will touch on the theoretical foundations of the material, our primary focus will be on getting our hands dirty by using the techniques covered to build models. The material covered in this class will be taught using real-world problems and place a high value on using messy, often-incomplete real-world data where the strengths and weaknesses of various tools can be evaluated. Prerequisites: (36-207 or 36-200 or 36-225 or 70-207 ) and ( 19-351 or 19-301 or 88-223 or 70-257 )