A friend of mine has recently started working on R-studio and is interested in filling the NA values in different columns using the above-mentioned function. Also, since he intends to run a time series analysis for every column, what should be the correct approach?
1 Answer
To replace by column means, an easy approach would be to use the base R function colMeans
. Let's say you have a data frame df
.
1) If you want to replace the NAs per column one by one, you could try this:
df <- sapply(df, function(x)ifelse(is.na(x), mean(x, na.rm=TRUE), x))
2) If you want to replace all NAs in one go, you could try this:
df <- ifelse(is.na(df), rep(colMeans(df, na.rm=TRUE), rep(nrow(df), ncol(df))), unlist(df))
1.
Is imputation the right thing to do? Sometimes, replacing the missing values does not make sense.2.
If imputation is the right choice, what exactly do you want to achieve:a)
replace by column mean,b)
replace by row meanc)
or replace by the mean of a given user based on other responses they have $\endgroup$