« The absurdity of predictive policing | Main | How to outsmart the election pundits »


Feed You can follow this conversation by subscribing to the comment feed for this post.

David Nicholas

Mr. Fung, just to let you know, you have a slight misspell in your article. So as data roll in, we should keep an eye on the morality rate. Morality should be mortality. Respectfully,


DN: Thanks for the catch. Fixed!

Antonio Rinaldi

I've read an article which speculates that chinese authorities are underestimating the number of infected.
If this is true the real infection rate is higher than that reported but the real mortality rate is lower. It should be interesting to investigate the implications of such hypothesis.


AR: Yes, that's another good point. Some also speculate that deaths are being covered up. These theories are in essence predictions about what might happen to the cases outside China. So far I've seen nothing from those cases to believe the allegations.


The main problem with the deaths is that it takes a while for them to appear, as from first symptoms it typically takes a week for the severe symptoms to appear. Deaths may be an embarrassment to China's government as they are likely to be much higher in some areas indicating a less well funded health system.

One would hope that they are being honest about the infection numbers. The problem is that if they aren't then they cannot judge the effectiveness of the interventions. At the moment the interventions look like they are achieving something, but slowly. The test will be over the next fortnight, as people return to work.

If anyone is interested, there is now a paper on treatments and outcomes https://jamanetwork.com/journals/jama/fullarticle/2761044


The death rate is discussed at https://www.worldometers.info/coronavirus/coronavirus-death-rate/ A reasonable way of calculating the death rate is to use the naive estimate but with the cases at the current time less the average time to death, which we unfortunately don't have because the Chinese are keeping death details to themselves. So at current date we have (for China only) 1381 deaths and if we assume 7 days to death 31211 cases or a death rate of 4.4%. If instead we assume 10 days to death it will be 6.7%, which I predict will be closer to the truth. The Chinese presumably have the data, which they could do a quick Kaplan-Meier on, and get the results fairly quickly, but don't want it known.

The comments to this entry are closed.

Get new posts by email:
Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR, Wired.

See my Youtube and Flickr.


  • only in Big Data
Numbers Rule Your World:
Amazon - Barnes&Noble

Amazon - Barnes&Noble

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee

Next Events

Jan: 10 NYPL Data Science Careers Talk, New York, NY

Past Events

Aug: 15 NYPL Analytics Resume Review Workshop, New York, NY

Apr: 2 Data Visualization Seminar, Pasadena, CA

Mar: 30 ASA DataFest, New York, NY

See more here

Principal Analytics Prep

Link to Principal Analytics Prep