We are borrowing material from a short course on Simulation-Based Inference for Epidemiological Dynamics (SBIED). This short course uses the same notation as STATS/DATASCI 531 and shares an emphasis on likelihood methods for inference. Indeed, SBIED can be viewed as a sub-course within STATS/DATASCI 531. Although SBIED focuses on epidemiological examples, you can check from previous final projects for 2016, 2018 and 2020 that the ideas extend readily to a variety of other situations. Later, in Chapter 16, we will consider models for financial markets, which is another major area of application for nonlinear time series techniques.
Chapter 11 Part 2 has overlap with Chapter 10. It may be no bad thing to see the foundational material again, from a slightly different perspective. You are welcome to watch at 1.5 x speed if you are already comfortable with Chapter 10.
| Chapter 11, Part 1 | Time series analysis and mechanistic models in ecology and epidemiology | Lecture video (20 mins) |
| Chapter 11, Part 2 | Introduction to POMP models | Lecture video (22 mins) |
| Chapter 11, Part 3 | The pomp R package for POMP models | Lecture video (7 mins) |
| Slides | ||
| Notes |
Back to course homepage
Acknowledgements
Source code for these notes