Modern Design of Experiments

Description from DukeHub

With breakthroughs in scientific computing, complex phenomena can now be reliably simulated or experimented on. However, these experiments are often expensive, and a key challenge is the design of such experiments to facilitate scalable and timely decision making. This course introduces experimental design methods for physical and computer experiments, Bayesian sampling and optimization, A/B testing and multi-armed bandits, and big data analytics. Emphasis is placed on understanding methodology and implementation for practical applications. Students should be comfortable with mathematical statistics at the level of STA 532/732 and Bayesian modeling at the level of STA 601L/602L/531.

Skills

• data and statisticsdescription: "bayesian" and 5 more
• designdescription: "design" and 1 more
• research methodsdescription: "experimental design" and 1 more

AI-resilient: 1 of 5 1/5

• No: Physical, lab or field work
• No: Live work with people
• ✓Yes: Judgment someone is accountable for · "decision"
• No: A craft you practice with your hands or body
• No: Small, in-person format

The checks lean on Microsoft Research, Working with AI (2025), which found generative AI applies least to physical work and hands-on work with people. A signal, not a promise. How the score works

More in STA

• Data Analysis and Statistical Inference STA 101L
• Studying Incarceration with Data STA 116CNL
• Introduction to Data Science and Statistical Thinking STA 199CCL
• Regression Analysis STA 210L
• Regression Analysis: Theory and Applications STA 221L
• Statistical Computing STA 323L

All STA courses · All of Graduate School