Statistics 697TS - ST-Time Series Analysis & Appl

This course will cover several workhorse models for analysis of time series data. The course will begin with a thorough and careful review of linear and general linear regression models, with a focus on model selection and uncertainty quantification. Basic time series concepts will then be introduced. Having built a strong foundation to work from, we will delve into several foundational time series models: autoregressive and vector autoregressive models. We will then introduce the state-space modeling framework, which generalizes the foundational time series models and offers greater flexibility. Time series models are especially computationally challenging to work with - throughout the course we will explore and implement the specialized algorithms that make computation feasible in R and/or STAN. Weekly problem sets, two-to-three short exams, and a final project will be required.

Open to Graduate Math and Statistics students only. Prerequisites: STAT 607/608 for familiarity with maximum likelihood estimation. STAT 625 or 705 for familiarity with linear algebra, specifically in the context of regression, recommended but not required.