correlation function always NA

问题:

I'm trying to find the correlations with this function but I just get NA for each input:

corr <- function(directory, threshold = 0) {    filenames <- list.files(directory, pattern="*.csv", full.names=TRUE)   csv <- lapply(filenames, read.csv)    correlations <- c()   c(correlations, lapply(csv, function(x){     cor(x$nitrate,x$sulfate)      print(cor(x$nitrate,x$sulfate))   }))    correlations   } 

If I change the function to print the nitrate and sulfate it does in fact have values, if I just pass in x to the cor function it complains about non-numeric values.

Why does it always print NA and not the actual correlation?

回答1:

The function cor has an argument use, which controls how it deals with missing observations, i.e. NAs. You might try setting it to use = "complete.obs".

文章来源: correlation function always NA