Introduction

I have always liked making predictions using different types of mathematical approaches. Today, I made a short video to show how I am making estimates of when the the 1,000,000th confirmed case of the Coronavirus will occur.

The Johns Hopkins Coronavirus Dashboard

You can access the Johns Hopkins coronavirus dashboard with this link. Once or twice a day I capture an image of this dashboard to record the development of the Corona cases around the world. Data dorks like me enjoy doing these types of things. You can see about 26 screen captures in the slideshow shown below.

This slideshow requires JavaScript.

The Complete Johns Hopkins Data

I know some of you must be astounded that I would take such a novice approach of screen capturing the dashboards and typing in the data. That type of approach really helps you appreciate the changing nature of situation, so that is why I am doing it. It is nice to get back to the basics of scientific data collection and analysis every once in a while. Don’t fret, however, because I also do the complete job with all the data, too.

You can access the JH data on github from this link. Yes, I also have an Alteryx workflow to process this data. I’m not going to show what I do with it, however. I’ll save that for a future article. Here is a picture of my initial workflow.

My Alteryx workflow to process the JH coronavirus data available on github (https://github.com/CSSEGISandData/COVID-19)

The Video for Predicting the 1,000,000th Confirmed Case

If you want to learn how to easily make exponential growth predictions, watch this video.

If you want to learn more about exponential growth modeling, watch this video which includes information on using Tableau to do the modeling.

The Current Estimate of the Millionth Case

Although we are not quite half-way to 1,000,000 cases, the exponential growth model is predicting that the millionth case will occur early on April 2, 2020.

The exponential model predictions are shown in the following figure. I make a prediction about every two days to see how the worldwide model is changing over time.