vaweeklytests: Weekly VDH COVID-19 testing data by health district
In rplzzz/vdhcovid: COVID-19 data by health district from Virginia Department of Health
Description
Usage
Format
Details
Source
Description
VDH testing and positive case fraction reports aggregated to weekly resolution
and adjusted for overdispersion.
Usage
1
Format
Data frame with 10 columns
- HealthDistrict
VDH health district.
- week
Week number
- date
Date of the last day of the week with recorded measurements.
- ntest
Number of reported tests for the week.
- npos
Number of reported cases for the week.
- varpos
Variance in the daily positive test fraction.
- nday
Number of days reported in the week.
- neff
Effective number of samples (see details).
- nposeff
Effective number of positive samples.
Details
Test counts and reports are aggregated by week, with weeks running Monday
through Sunday (they are defined this way so that a Monday morning update
has data for a full week). The date assigned is the date of the last day (Sunday)
of the week.
The positive test fraction is
calculated as the number of positive tests for the week divided by the total
number of tests for the week.
To adjust for overdispersion we also calculate the daily positive test fraction
and its variance over the 7 days in the week. The effective number of samples
is calculated as p*(1-p)/sig2, where p is the weekly positive test fraction.
Loosely speaking, what we're doing here is assuming that the sample prevalence
is constant over a week and taking the variance of the daily measurements as
an estimate of the variance in the measurement procedure. Then we set the
effective number of tests such that the variance of the posterior PDF of the
p inferred from those observations will be equal to the observed variance.
From this, the effective number of positive results is estimated by multiplying
neff by fpos. Both neff and nposeff are rounded to the nearest integer (meaning
that nposeff/neff may not be precisely equal fpos).
Finally, note that this procedure is performed only when the number of observations
for the week is greater than 32 and there are at least 4 days of observations
in the week. Weeks that don't meet these standards in a district are left
unadjusted. This prevents us from getting crazy results for partial weeks or
weeks with just a few observations. We also force neff <= ntest (so far only
one week in one district has required adjustment in this respect).
Source
Virginia Department of Health
https://www.vdh.virginia.gov/coronavirus/
rplzzz/vdhcovid documentation built on July 2, 2021, 7:43 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description Usage Format Details Source
Description
VDH testing and positive case fraction reports aggregated to weekly resolution and adjusted for overdispersion.
Usage
1 |
Format
Data frame with 10 columns
- HealthDistrict
VDH health district.
- week
Week number
- date
Date of the last day of the week with recorded measurements.
- ntest
Number of reported tests for the week.
- npos
Number of reported cases for the week.
- varpos
Variance in the daily positive test fraction.
- nday
Number of days reported in the week.
- neff
Effective number of samples (see details).
- nposeff
Effective number of positive samples.
Details
Test counts and reports are aggregated by week, with weeks running Monday through Sunday (they are defined this way so that a Monday morning update has data for a full week). The date assigned is the date of the last day (Sunday) of the week. The positive test fraction is calculated as the number of positive tests for the week divided by the total number of tests for the week.
To adjust for overdispersion we also calculate the daily positive test fraction and its variance over the 7 days in the week. The effective number of samples is calculated as p*(1-p)/sig2, where p is the weekly positive test fraction. Loosely speaking, what we're doing here is assuming that the sample prevalence is constant over a week and taking the variance of the daily measurements as an estimate of the variance in the measurement procedure. Then we set the effective number of tests such that the variance of the posterior PDF of the p inferred from those observations will be equal to the observed variance. From this, the effective number of positive results is estimated by multiplying neff by fpos. Both neff and nposeff are rounded to the nearest integer (meaning that nposeff/neff may not be precisely equal fpos).
Finally, note that this procedure is performed only when the number of observations for the week is greater than 32 and there are at least 4 days of observations in the week. Weeks that don't meet these standards in a district are left unadjusted. This prevents us from getting crazy results for partial weeks or weeks with just a few observations. We also force neff <= ntest (so far only one week in one district has required adjustment in this respect).
Source
Virginia Department of Health https://www.vdh.virginia.gov/coronavirus/
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.