statistics Project

Final Project: Modeling a Pandemic

Due Friday, December 4, 2020

In this final project, we will look at the application of mathematical and statistical

modeling applied to understanding the behavior of pandemics and improving on public

health outcomes. A statistical model of a pandemic can address many important

questions.

 If a new disease emerges, is the initial spread of disease inevitable? What

factors affect the probability that it will infect a critical mass of individuals?

 Once a disease is established in a critical mass of individuals, how rapidly does it

spread? What factors/variables control the rate that the disease can spread to

different parts of a population?

 If some parts of the population are disproportionately affected, how does that

affect the control strategies?

 How does variability in the range of symptoms (asymptomatic, mild, severe)

affect the control strategies?

 Why were some past epidemics easy to control and others like COVID-19 more

difficult?

In your report, you will want to describe how to build a statistical model to address one

of the above questions. You should make sure to address the following features of your

statistical model.

 Design: What are the features of your statistical model and what are the

important parameters?

 Estimation: How would you collect data to estimate key parameters of the

model? Consider how to address potential sources of bias. A fitted statistical

model is only useful if the underlying parameter estimates are reasonably

accurate.

 Prediction: What is the model trying to predict and how can this be useful?

 Sensitivity Analysis: Which parameters are the model outcomes most sensitive

to? Which parameter values can be modified and how? For example, the

infectivity rate per person can be reduced to zero by quarantining everyone and

obtaining perfect compliance.

There are several options in how you can approach your final report. The important

aspect of your report is clearly describing a statistical model, and your subsequent

discussion of the model. You should be sure to cite three articles or well-researched

news sources (not just a short blurb on a news website). Discuss these articles in light

of the statistical model you chose to focus on. For example, if you chose a statistical

model to predict how fast a disease spreads, you could compare diseases like COVID-

19 to MERS or H1N1 virus. Which parameter values were different, so that these other

diseases did not spread as rapidly? Another option would be to focus on the challenges

in obtaining data and accurate estimates for the parameters in your model.

You are welcome to include your own simulations in R, although this is optional. Your

final report should be 3-5 pages in length, including any graphs and/or tables. You can

work alone or in groups of up to three as long as you practice social distancing and

communicate with each other remotely. Please communicate with me (in the next week

or so) and let me know: your topic, group members, articles you plan to cite, and any

questions you may have.