Data Science Decisions in Time: Using Data Effectively

This module introduces simple sequential data and Wald's model where data arrives from a distribution and is not time dependent. Covers generative data, health and business data collection, and connections to code work and AI.

This module bridges into Markov Processes and Markov Chains. Covers Thompson sampling algorithm, its applications in A:B testing and risk assessment for neglected tropical diseases, including spatial dimension considerations.

Change points are locations where previously stationary distributions shift to a new distribution. Covers applications in manufacturing, climate change, and uses Gaussian models for change point determination.

Markov chains describe sequences of state changes in complex systems. Module covers modeling transitions between states, failure rates in bridges and buildings, and matrix operations for chain analysis.

Advanced modeling of Markov processes with decisions, connecting to reinforcement learning. Covers optimization of decision sets, parameterization of processes, with applications in emergency department resource management and drug treatment planning.