Hello I am trying to loop over a list of dataframes and pull samples of differing size from the dataframes. For example for df1, I want a sample of size 10, df2 a sample of size 8, etc. I have worked with Melissa Key on my previously asked question to develop the following code:
sampler <- function(df, n,...) {
return(df[sample(x=nrow(df),n),])
}
#for(i in 1:totalstratum){
sample_list<-lapply(population_list,sampler,n=stratum_sizes[i,1])
#}
#or
library(purrr)
sample_list<-map2(population_list, stratum_sizes,sampler)
where stratum_sizes is a vector {4,5,3,2,10,10,8} and totalstratum=nrow(stratum_sizes), which is also equal to the number of elements in the list population_list.
So far, I am able to get a sample, but never with the correct number of observations. Any ideas? Thank you in advance for any help!
I assume you'd like to sample a certain number of rows from data.frames stored in a list.
How about the following using map2:
# Generate sample data
# Here: A list of three data.frames
set.seed(2017);
lst <- lapply(1:3, function(x) data.frame(val1 = runif(20), val2 = runif(20)))
# Define the sample sizes for every data.frame in the list
ssize <- c(10, 5, 3)
# Sample ssize entries from every data.frame in the list
map2(ssize, lst, ~ .y[sample(nrow(.y), .x), ])
#[[1]]
# val1 val2
#16 0.38868193 0.6500038
#8 0.43490560 0.3191046
#11 0.67433148 0.8838444
#7 0.03932234 0.6204450
#2 0.53717641 0.3798674
#3 0.46919565 0.9420740
#19 0.94099988 0.1771317
#5 0.77008816 0.2276118
#10 0.27383312 0.2608393
#14 0.43207779 0.2117630
#
#[[2]]
# val1 val2
#12 0.8835366 0.6904628
#4 0.0791699 0.7512366
#6 0.5096950 0.4699963
#19 0.5393251 0.4123170
#20 0.9229542 0.9327490
#
#[[3]]
# val1 val2
#4 0.9204118 0.1926415
#15 0.8373573 0.9309950
#8 0.1653395 0.5895154