Epid 814 - Advanced Modeling Methods

Course Materials

Organized by type:

• Readings

• Lectures

• Labs

---

Course schedule

Topics are subject to change and rearrangement!

1. Compartmental Models and DAGs

• Compartmental Models and DAGs (Josh Havumaki) - Discussion paper

2. Parameter Estimation and Maximum Likelihood

• Introduction to Parameter Estimation and Maximum Likelihood - Lecture

• Parameter Estimation Coding Session - Lab Assignment

• Extra mini-lecture by special request - Separation and Firth’s Penalized Likelihood

3. Uncertainty and Identifiability

• Identifiability and Parameter Uncertainty - Lecture

• Identifiability and Uncertainty Lab

4. Visualization with ggplot2

• Lab - our collection of ggplot examples

• Nice, comprehensive ggplot tutorial

• Another tutorial, and an introductory course on datacamp

• Mapping with ggplot

5. MCMC & Bayesian approaches

• Lecture - Bayesian Approaches to Parameter Estimation

• Lab - Metropolis Sampler

6. R0 & dynamical systems/threshold conditions (including stochastic R0, type/target R0’s)

• Lecture - Understanding the Basic Reproduction Number (Andrew Brouwer)

• Reading - A linear algebraic approach to understanding the basic reproduction number (Andrew Brouwer)

5.5 A little more Bayesian stuff (see Section 5 also)

• BayesianTools Lab

• And code from Alicia for pulling a quantile from a sample (e.g. to generate credible intervals, etc)

7. Latin Hypercube Sampling

• Lecture - Latin Hypercube Sampling (Rachel Gicquelais)

• LHS Example Lab

• Readings

  • Marino et al., 2008

  • Morris and Mitchell, 1995

8. Data scraping and general wrangling (e.g. twitter scraping, MMWR scraping)

• Using R to pull data using API’s - Twitter, Facebook, and Google Maps (Yu-Han Kao)
  • Data Scraping Lab Part 1
    
    

• Data visualization using D3

  • Part 1: SVG, JavaScript, and D3 Lab

  • Finished example file

  • Part 2: Network Visualization with JavaScript and D3
    
    

• General web-scraping (e.g. maybe MMWR, Promed/Healthmap, Wikipedia, Pubmed etc?)

9. Stochastic processes and stochastic modeling methods

Maybe include

• Gillespie & Tau-leaping (Alicia?)

10. Additional Topics

• Spatial modeling & spatial stats

• Network models - particularly integrating statistics with network models and data

• Working with sequencing data