corr <- function(directory, threshold = 0) {
# Get full path of the specsdata folder.
directory <- paste(getwd(), "/", directory, "/", sep = "")
# Initialize data frame.
correlations <- numeric()
# Get list of files.
files <- list.files(directory)
# For each file:
for (file in files) {
# Read the file.
data <- read.csv(paste(directory, file, sep = ""))
# Count the number of complete cases.
nobs <- sum(complete.cases(data))
# If the number of complete cases is greater than the threshold:
if (nobs > threshold) {
# Calculate the correlation between sulfate and nitrate.
correlation <- cor(data$sulfate, data$nitrate)
# Add the correlation to the vector of correlations.
correlations <- c(correlations, correlation)
}
}
# Return the vector of correlations.
return(correlations)
}
This function takes two arguments: `directory` and `threshold`. The `directory` argument specifies the directory where the data files are located. The `threshold` argument specifies the minimum number of complete cases required to calculate the correlation.
The function first gets the full path of the `specdata` folder. Then, it initializes a data frame to store the correlations. For each file in the directory, the function reads the file, counts the number of complete cases, and calculates the correlation between sulfate and nitrate. If the number of complete cases is greater than the threshold, the correlation is added to the data frame. Finally, the function returns the data frame.
Here is an example output from the `corr` function:
corr("specdata", 150)
# [1] -0.01896 -0.14051 -0.04390 -0.06816 -0.12351 -0.07589
