Team Toad: Data Groupies About

Team Toad Data Groupies - Skew-Normal Epidemic Modeling

Introduction Graphs Blog Theory About Contact

About Michael Mauldin AKA "Fuzzy"

Dr. Michael L Mauldin, known as "Fuzzy" to his friends, is a retired computer scientist who majored in mathematical sciences at Rice University (BA MathSci 1981, Sid Rich). He was a teaching assistant for several courses, including Statistics.

His master's thesis Maintaining Diversity in Genetic Search (1984) formed part of the code for the curve fitter used in this epidemic model.

His PhD thesis and subsequent work on natural language and information retrieval are irrelevant to this epidemic modeling project (see Fuzzy's vita).

Disclaimer: none of this has been peer-reviewed.

My credentials allow me to make statements about math, but not medicine.
My goal is to explain the math to you so that you can draw your own
conclusions about the state of the public heath threat.

About Team Toad

Team Toad is known primarily for building and competing with fighting robots. We also do Jeeping and offroad camping. The team consists of Fuzzy, his wife Debbie, and their family and friends who've joined in our activities.

The team itself has no special medical or epidemiological expertise.

Data Sources

These are the sources for both the historical medical data on confirmed cases and deaths in the United States and the World. We believe these data to be reliable, and make no recommendation about whose data source is better or worse than any other.

Johns Hopkins University was first out of the gate with their viral pandemic tracker (JHU Coronavirus Resource Center). This was our only data source for the first few months of 2020.

The New York Times formed a task force to collect and distribute case and death information about the coronavirus on March 27, 2020 (NY Times GitHub), and this quickly became our primary source of data for the United States. Kudos and thanks to the NY Times Coronavirus team for their outstanding work. Their data is licensed for non-commercial use, with attribution: NYT U.S. tracking page

On June 18, 2021 I switched to using rolling averages computed by the NY Times, because their raw data had an increasing number of anomalies that became difficult to reconcile. Charts from that day forward do not show outliers separately.

I started using the NYT Coronavirus Variant tracker for percentages of the variants in the US.

Center for Disease Control (CDC)

The CDC publishes a Variant Proportions web page.

The CDC also provides the CDC Coronavirus Vaccine tracker for reports on how many US residents have been vaccinated.

The Oxford Martin School has a resource page called OurWorldInData.org that also has a GitHub page to share data about the coronavirus throughout the world. This has become our source of data for our "World ex-China" graphs.

Worldometer provides a incredibly rich and interactive web site for tracking the epidemic at Worldometers.info Authors: Hannah Ritchie, Esteban Ortiz-Ospina, Diana Beltekian, Edouard Mathieu, Joe Hasell, Bobbie Macdonald, Charlie Giattino, and Max Roser. Web development: Breck Yunits, Ernst van Woerden, Daniel Gavrilov, Matthieu Bergel, Shahid Ahmad, and Jason Crawford.

References

For information on the skew-normal curve, I used Wikipedia's entry on the skew-normal distribution.

For general background information and terminology I used Gordis Epidemiology, Sixth Ed. (2019) by David D. Celentano and Myses Szklo, both at Johns Hopkins School of Public Health.

Resources

I wrote the curve fitter programs in C, just C, the way God and Dennis Ritchie intended.

I wrote several data munging scripts in Perl to prepare the various CSV spreadsheets for both the curve fitter and for making graphs.

All modeling runs were made on a System76 Galago Pro (2019) running Manjaro Linux.

Spreadsheets and graphs by LibreOffice Calc.

Map data was processed using QGis 3.10 with shapefiles from the US Census Bureau and the World Borders Dataset provided by Bjorn Sandvik at thematicmapping.org.