bp10k {clustra} | R Documentation |
Simulated blood pressure data
Description
A sample of 10,000 individuals from the full 80,000 individuals in a dataset available on GitHub at https://github.com/MVP-CHAMPION/clustra-SAS/raw/main/bp_data/simulated_data_27June2023.csv.gz
Usage
bp10k
Format
bp10k
A "data.table" and "data.frame" with 167,277 rows and 4 columns:
- id
An integer in 1:80000.
- group
An integer in 1:5.
- time
An integer between -365 and 730, giving observation day with reference to an intervention at time 0.
- response
The systolic blood pressure on that day.
Details
The full data set contains 80,000 individuals, each with an average of about 17 observations in 5 clusters with scatter. The individuals are generated from a 5-cluster thin spline model of actual blood pressures collected from roughly the same number of individuals at U.S. Department of Veterans Affairs facilities in connection with the MVP-CHAMPION project. Each cluster-mean generated individual has a random number of observations at random times with one observation at intervention time 0, and with added standard normal error. The resulting data has 1,353,910 rows and 4 columns.