Custom Function to replace missing values in a vector with the mean of values

Question

I have this vector:

a<-c(1,2,3,NA,4,5,NA,NA,100,45,NA)

and I'm trying to replace all the missing values present in this vector by writing this custom function:

mean_impute<-function(x){
ifelse(is.na(x),mean(x),x)
}

But the result obtained has NA values in it. What is wrong with this function?

Answer 1

You have missed out on "na.rm=TRUE" inside the mean function.

So, your should look like this:

mean_impute<-function(x){
ifelse(is.na(x),mean(x,na.rm = T),x)
}

And this code will give you the desired output which is:

1.00000   2.00000   3.00000  22.85714   4.00000   5.00000  22.85714  22.85714 100.00000  45.00000  22.85714

Answer 2

Try this.

lapply(a,function(x){ifelse(is.na(x),mean(a,na.rm = TRUE),x) })