My problem: a large dataset with columns containing very few values.
What I whish: reordering columns with sorting frequency of missing values condition.
What I tried:
# my data.frame
data <- data.frame(column1 = c(NA,NA,NA,NA), # 4 NA
column2 = c(4, NA, NA, 1), # 2 NA
column3 = c(12, NA, 5, 78), # 1 NA
column4 = c(1, 2, 13, 4)) # NO NA
# relative frequency of NA values
missing_values <- colMeans(is.na(data))
# What I want:
column4, column3, column2, column1
