I spent a few days exploring various models hoping to get a sense of where this coronavirus is heading. Since I was planning a trip back to Atlanta, and I hear all this news about a surge of infection and I can see that myself, so I was trying to see if there is a maximum and how long will it last.
The short answer is no one knows. I think when I arrive in Atlanta it might be the peak of the crisis.
Things are not looking good in Atlanta. They are having 4000 infections a day in Georgia, and it is possible they might get to 10,000 like in Florida and Texas in next couple weeks. (Made me wonder on Georgia, Florida, Texas, or Arizona’s reopening stategy)
10,000 is high. I don’t think will ever get to 20,000. I finally understand the magnitude of the number. I was before thinking hey we have a million people in a city and 10x that in a state, 10,000 is just a fraction (1%, or 0.1%). Many people were saying the news are over hyping the severity of the pandemic.
I almost attempted to do my own forecast but I found many models were already available on the web. One was the SIR. (there is a website from someone making this very easy).
It divides the whole population into three groups, the Susceptible, the Infected and the Recovered.
An observation I had, was most models focus on the initial weeks of the spread where the infection point is small (like 1-1000). The time frame was usually the first 4-8 weeks.
I need a model where we are in the midst of the pandemic, where the infection is every where! And it is no long weeks, we are into months.
Another observation is most models (99%) expect to reach a steady state after a time, that is the infection would die out! This might be true with most outbreaks. Did they say there could be a second wave?
The total of infections/deaths curve when graphed would be an s curve (logistic curve). And the rate of infection would be a bell shape.
Seeing some S shape curves in a few states make me to believe too the pandemic was over.
Something I didn’t understand when I took the probability class in college when the professor kept talking about density function and I kept thinking what the heck it has to do with density.
Suddenly after couple decades, it connected. I was dividing the rate of infection by the population and graphing it. It came to me this is a density curve. I had an ureka moment. This stuff is related to probability. You can calculate the sum of the area under the curve to find the probability of an event (I used to hate words like that: Event, Random Variable, Function)
Summary/out take. I stopped my work. I am too lazy to pull data and do the math to carry it to the end. It takes too much time and effort. Playing around with different models I had some learning experience and it was useful. The practical aspect is I might be arriving in Atlanta at the peak of things this time around. Does this change my mind about taking my trip? Probably not.
Morality side: there was a sense of ‘righteous’ anger from people when they saw certain segment of the society blatantly ignoring social distancing by going to bars and beaches in the middle of a pandemic just for their pleasure, making the pandemic worse for everyone. I do too, so I have to balance out whether to travel or stay at home. (I heard this on NPR on Friday morning jul 24, also social shaming might not work so well).