Simulations

The following simulations span over three months and consider different factors such as age and mortality rate to create a realistic depiction of the spread of COVID-19. The simulations illustrate the effect that contact tracing has on the spread of the virus. Two types of models were created, one with unrealistic clustering but realistic degree distribution (Barabasi-Albert Graph) and one with realistic clustering but unrealistic degree distribution (Watts-Strogatz Graph). The links below contain more information about each model's simulations.

Code can be found at https://github.com/Eimara/COVID/tree/master

Participation in Contact Tracing

The following simulations overview the differences in fatality rate depending on what percent of the population agree to participate in contact tracing.

Barabasi-Albert Model

Because this model uses realistic degree distribution, it is able to include individuals who come into contact with more people than others individuals do. For example, an individual working as a store clerk will likely meet more people than someone who stays at home more often.

more info

Watts-Strogatz Model

Because this model uses realistic clustering, it is able to show how people form smaller social circles with each other.

more info

The graphs were constructed using an age distribution of New York City's population in 2018. Individuals were categorized into groups based on risk :

• very low risk with less than 1% mortality rate

• low risk with 1% mortality rate

• moderate risk with 3% mortality rate

• high risk with 10% mortality rate

These numbers are subject to change based on new data

In a perfect situation, self-quarantine measures taken early on will result in 0 total deaths given the current infection and mortality rates. If self-quarantine is delayed, results will differ.