How to impute missing values with row mean in R

问题

From a large data frame, I have extracted a row of numeric data and saved as a vector. Some of the values are missing and marked as NA. I want to impute the missing values with row mean.

Thanks

回答1:

Let x be your vector:

x <- c(NA,0,2,0,2,NA,NA,NA,0,2)
ifelse(is.na(x), mean(x, na.rm = TRUE), x)
# [1] 1 0 2 0 2 1 1 1 0 2

Or if you don't care for the original vector, you can modify it directly:

x[is.na(x)] <- mean(x, na.rm = TRUE)

回答2:

Use this:

filter <- is.na(myVec)

myVec[filter] <- colMeans(myDF[,filter], na.rm=TRUE)

Where myVec is your vector and myDF is your data.frame.

来源：https://stackoverflow.com/questions/17203955/how-to-impute-missing-values-with-row-mean-in-r