531w26

Modeling and Analysis of Time Series Data (STATS 531)

Chapter 13: Simulation of stochastic dynamic models

As for Chapter 12, we use material from the short course Simulation-Based Inference for Epidemiological Dynamics (SBIED). We develop a concrete example of the general POMP modeling framework, and we see the theory and practice of implementing a simulator for the model.

The Susceptible-Infected-Recovered (SIR) model used for this chapter is a central concept for epidemiology. For the purposes of STATS 531, we view it as one example of a mechanistic model, which exemplifies a more general process of model development and data analysis. One epidemiological idea used without definition in the lecture is $R_0$, defined to be the expected number of secondary infections arising from one infected individual in a fully susceptible population. The SIR model supposes that previously infected individuals cannot become reinfected, so those in compartment R are protected from infection.

The SIR model in epidemiology is closely related to predator-prey models in ecology. Similar models can be used to describe spread of ideas (including rumors or mis-information) on social networks.

Lecture material Link
Slides pdf
Notes pdf
Recording, Summer 2021, Part 1 (17 mins)
Recording, Summer 2021, Part 2 (24 mins)
Recording, Summer 2021, Part 3 (47 mins)
Recording, Summer 2021, Part 4 (10 mins)

Back to course homepage
Acknowledgements
Source code for these notes