144
Beginner’s Guide to Code Algorithms
144
11.1 FEATURES OF THE GRAPH
Let’s begin with the chart and then look at the other different features that are put
together around it. Best done through an example, here is what we want.
A graph of number of COVID cases by US state (once you see the technique, you
can easily extend it to other geographic regions around the world). We want it by day
because that shows a clear trend (upward or downward). We want to see cumulative
up to each date, as well as new cases separately because each means slightly different
things in the trend chart. While most rely on this metric to determine how safe a place is,
the other metric is number of deaths due to COVID. Deaths typically lag the infection
by several weeks to several months, hence may not be the best indicator of safety, but
certainly an indicator of other things such as the state’s ability to handle cases, quality of
care, and so on. It would be great to see comparisons of different states. Also, we would
like to see the top N states so that that gives us a sense for the worst affected states.
We must remember that numbers are not everything—there are many factors in
volved in judging safety, and this analysis is just an experiment to help you develop
charts. However, charts and graphs have a pretty loudmouth and tell a story that is
worth a thousand words.
11.2 THE DATA
Before we begin, we have to get the data. Luckily, COVID data are publicly available
from multiple sources and are updated on a daily basis. The source I have chosen
for this exercise is https://usafactsstatic.blob.core.windows.net/public/data/covid-19.
There are two files that are available in this web address:
• covid_confirmed_usafacts.csv
• covid_deaths_usafacts.csv
The names are self-explanatory.
Both these files have the same format as shown below, each cell showing the
number of new cases in the days specified in the column header.
FIGURE 11.6 covid_confirmed_usafacts.csv.