Applied Forecasting in Complex Systems ((4 Credits))

Catalog description

This course introduces forecasting methods for complex and dynamic systems such as transportation flows, supply chains, power networks, and public health systems. Learners build and compare statistical and machine-learning forecasting models, quantify uncertainty, and evaluate forecast performance for decision-making. Topics include time-series decomposition, regression-based forecasting, autoregressive integrated moving average (ARIMA) and seasonal ARIMA models, intervention and state-space models, multivariate forecasting, backtesting, and probabilistic forecasting. Using Python-based tools and authentic case studies, learners work with real time-series data, diagnose model performance, and communicate forecast results in technical reports designed for operational and strategic use. Prerequisites: MATH 3060 Mathematical Statistics, CSCI 2020 Introduction to Programming with Python, and MATH 3010 Linear Algebra or MATH 3011 Applied Linear Algebra.