search

CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE
https://systems.jhu.edu/research/public-health/ncov/
29.11k stars 18.39k forks source link

April 5 to 6 US change rate #2108

Open hdoran opened 4 years ago

hdoran commented 4 years ago

The time series data shows a day to day increase in the US death count of 372 from April 5 to 6, which differs from the previously reported data and deviates largely from the day-to-day trend rate. Is this change accurate

April 4 to 5 change: 1279 April 5 to 6 change: 372 April 6 to 7 change: 1938 April 7 to 8 change: 1973 April 8 to 9 change: 1783

hdoran commented 4 years ago

R code to replicate issue reported dat <- read.csv('https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_deaths_US.csv')

Daily death rate

diff(sapply(13:ncol(dat), function(i) sum(dat[,i])))

Increase from prior day death rate

diff(diff(sapply(13:ncol(dat), function(i) sum(dat[,i]))))

paolinic03 commented 4 years ago

See response on top discussion here regarding NY death inconsistencies. Not sure it'll help but at least you get an answer on potential fix timeline.

https://community.tableau.com/community/resources/covid-19-forum/activity?_ga=2.79888337.753758224.1586104679-108680942.1584906703