Disclaimer
I still am not an expert on medicine or statistics. All graphs are no better than the data they are based on, and I don't know how good the data is. Past performance does not always predict future direction of little squiggly lines on graphs.
Today
Today, I have just updated existing graphs. The situation inside China, as far as the official numbers go, continues to improve. The situation outside China appears to be getting worse.
The graphs
Figure 1. This graph is based on the "confirmed" reports from various countries health departments, minus the dead and recovered. It seems to show the number of active infections peaking, then dropping a bit lately. It also, unfortunately, show a spike on February 12th, a statistical artifact based on a redefinition of "confirmed" by China on that date.
Figure 2. One way to "control" for the February 12th spike is to add in the count of "suspected" infections, because the redefinition appears have hinged on the exact borderline between suspected and confirmed. However, there is still some chaotic behavior here around the date of redefinition.
Figure 3. Another way to adjust for the spike of February 12th, which a bunch of prior "clinical diagnoses" on one day of reporting. Here we speculatively "redistribute" those clinical diagnoses across prior days. To avoid any fancy data manipulation, they are distributed evenly in proportion to the number of prior lab-tested cases by multiplying all active cases prior to February 12th by 1.2.
Figure 4. Here we see, ever since February 4th if you ignore the redefinition, or if you don't ever since February 12th, that the net movement in confirmed current infections is headed downhill, which is what we'd like to see.
Figure 5. If we toss in suspected infections on top of confirmed, we continue to see a similar decrease lately.
Figure 6. This is the percentage of total confirmed cases (not adjusted for deaths and recoveries) that are outside of China. The last eight days reflect a spike in the extent to which the virus has been found outside China.
Figure 7. This is the same as Figure 6, but if you remove the Princess Diamond cases. It does show that parts of the recent uptick is to some degree, but not entirely, driven by the happenings on a single cruise ship.
This page is released under the CC0 1.0 license.