For some reason, the more basic the action I want to do in R, the harder it seems to me… So, here’s how to replace NA values with any value so that I don’t have to look for this again.
First, the easy one: in a vector:
arff=1:10; arff[3]=NA; arff[which(is.na(arff))]=-1; arff; [1] 1 2 -1 4 5 6 7 8 9 1
In a matrix, oddly enough the syntax… is exactly the same!
arff=matrix(0,4,3);
arff[2,3]=NA;
arff[which(is.na(arff))]=5;
arff;
[,1] [,2] [,3]
[1,] 0 0 0
[2,] 0 0 5
[3,] 0 0 0
[4,] 0 0 0
Actually, not that odd if you consider the matrix as a 2D array the way it’s built in C++ or such. Still, that came as a surprise to me.
Now in a data frame:
pcr=matrix(0,4,3); pcr[2,3]=NA; pcr=data.frame(pcr); pcr=replace(pcr,is.na(pcr),5); pcr; X1 X2 X3 1 0 0 0 2 0 0 5 3 0 0 0 4 0 0 0
Actually, replace() works on matrices and vectors too 😉
Main source: R help – How to replace all
0 Responses
Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.