R: Filling Missing Values (NA) by Multiplying Two Separate Vectors

Question

I'm having a brain-freeze.

This is what I have:

C <- c(C1, C2, C3)  # A constant for every row in the data frame
r <- c(r1, r2, r3, r4)   # A ratio for every column in the data frame

My data frame looks like this:

     1    2    3   4
a   0.7  0.4   NA  NA
b    NA   NA  0.3  NA
c    NA  0.6   NA 0.4

I need to fill in the NA's with a multiplication of C and r so that it looks like this:

      1      2     3        4
a    0.7    0.4   C1*r3   C1*r4
b   C2*r1  C2*r2   0.3    C2*r4
c   C3*r1   0.6   C3*r3    0.4

Notice that the multiplication is only done for the NA's and not for numbers that already exist. I know is.na is used to pick out the NA's, and it's probably just linear algebra, but my brain has quit for the day. Any help would be great.

Thanks.

Answer 1

If mm is your matrix , you can fill missing values like this:

mm[is.na(mm)] <- outer(C,r)[is.na(mm)]

example with data :

mm <- read.table(text='   1    2    3   4
a   0.7  0.4   NA  NA
b    NA   NA  0.3  NA
c    NA  0.6   NA 0.4')

C <- c(1, 1, 1)  # A constant for every row in the data frame
r <- c(2, 2, 2, 2) 

mm[is.na(mm)] <- outer(C,r)[is.na(mm)]

#    X1  X2  X3  X4
# a 0.7 0.4 2.0 2.0
# b 2.0 2.0 0.3 2.0
# c 2.0 0.6 2.0 0.4