Models, Data and Inference for Socio-Technical Systems
MIT's Engineering Systems Division offers this graduate-level course on building probabilistic and statistical models of complex socio-technical systems. Students review and extend functions of random variables, Poisson processes, and Markov processes, then move into classical statistics: chi-squared, t, and f tests derived as functions of random variables, hypothesis testing, regression, correlation versus causation, basic data mining, and a comparison of Bayesian and classical statistical approaches. The course requires a class project applying these modeling techniques to a real system design or decision-making problem. Materials available through MIT OpenCourseWare include lecture notes and assignments, consistent with OCW's standard free, self-paced, no-certificate format. It suits students who already have some background in probability and want to apply statistical inference to engineering and organizational systems rather than pure theory.