An exploration of infection trends from Our World in Data, retrieved 1900 Pacific Time 22 March 2020. This data was still current on the website at press time, noon Pacific 23 March. The analytic community is currently awash in data about COVID-19, and it is at this time uncertain what its’eventual impact on the United States (and world) will be. In this short piece, we present two graphs which are derived from the data; these are slightly different than the ones shown on the website. For this article rendered with R language code in-line, please click here
Of initial interest is the relationship between testing and data. At first blush, it appears that the more you test, the more you find. It appears that the more you test, the more you find:
Relationship between Testing and Cases Discovered. At press time, the United States is in the lower left hand corner, with low numbers of cases and low numbers of tests. As time progresses, the key question is how the US will move up and to the right on this chart
Testing is a complicated process which we will not exlore here, other than to note that it is not a ‘random sample’ in the statistical sense; for example, there are individuals who will seek out a test when they have little cause to believe that they are infected, as well as conversely, individuals who will believe that they are not infected despite compelling evidence to the contrary.
| term | estimate | std.error | statistic | p.value |
|---|---|---|---|---|
| TestsPerMillion | 0.0372302 | 0.002563 | 14.52623 | 0 |
Worldwide, we see about .03 cases per test.
Infection Curve - China, Italy, Iran, USA
Coronavirus Scases per million for five selected nations
Here we accept the China data without comment. It appears that at this moment, the United states is at a critical point; will our countermeasures be effective and level off (like the reported China data), or will it experience exponential growth, like France, Italy, and Iran?