covidatlas / li

Next-generation serverless crawler for COVID-19 data
Apache License 2.0
57 stars 33 forks source link

The sum of cases for some states at county level does not match the value at state level #609

Open jingjtang opened 3 years ago

jingjtang commented 3 years ago

I noticed that the sum of cases for some state at county level does not match the value at state level. For example, if we use the county level data and aggregate them to state level, we would get such plots(unassigned cases has not been added) Compare_cumulative_between_CDS_and_JHU.pdf

However, if we plot the state level data directly and add the unassigned cases, we would get these, UseStateLevel.pdf

How did you get the number of unassigned cases? Are the assigned cases already included in the number of cases for states or not?

These are all of the unassigned cases in US according to timeseries.csv, it seems the cases does not match well with the states especially between Arkansas and Georgia. unassigned.xlsx