We continue using material from the short course Simulation-Based Inference for Epidemiological Dynamics (SBIED). Chapter 14 is Lesson 4 of SBIED. The main topic is likelihood maximization via an iterated particle filter. This enables a range of tools of likelihood-based inference to be applied—maximum likelihood estimation, likelihood ratio tests, profile likelihood confidence intervals, and AIC for model selection. Methods are demonstrated on a model for measles, but these techniques apply to the wide range of POMP models for which particle filtering is applicable.

Slides pdf
Annotated slides pdf
Notes pdf
R script R
Recording, Chapter 14, Sections I,II Classification of inference methods for POMP models (16 mins)
Recording, Chapter 14, Section III Introduction to iterated filtering (15 mins)
Recording, Chapter 14, Section IV Iterated filtering in practice (25 mins)
Recording, Chapter 14, Sections V Global likelihood maximization and profile likelihood (33 mins)
Recording, Chapter 14, Section VI Using likelihood and profile calculations to develop a data analysis (20 mins)


Back to course homepage
Acknowledgements
Source code for these notes