Question

I have a problem with my loop. It just delete some rows that have 0 or NA values in my desire column and I don’t know why:

for (i in 1:105) {
  for (j in 1:l[i+1]){
    if(m[[i]][j,12]==0 | is.na(m[[i]][j,12])) {
      m[[i]]=m[[i]][-j,]
    } 
  }
}

Searching on the web I saw that maybe I could use apply function… something like:

for( i in 1:105){m[[i]]<-m[[i]][!apply(is.na(m[[i]]), 1, any),]}

for( i in 1:105){
  as.null(0)
  m[[i]]<-m[[i]][!apply(is.null(m[[i]]), 1, any),] 
}

This throws me a dim(x) error… I want to set Zero number as NULL

I was thinking something as follows but clearly it isn’t good… it just the idea…. I really don’t know how to use apply function well

for( i in 1:105){as.null(0) m[[i]]<-!apply(m[[i]],1,is.null(m[[i]])) }

Thanks a lot for your useful help !

Answer 1

You use apply to apply a function over a margin of an array, but I think is not the best idea here, since you only need to subset the matrix properly. Let’s focus in just one matrix m.

ind = m[,12] == 0 | is.na(m[,12])

ind will have TRUE where appropiate and the you can do

m = m[!ind, ] # m is a matrix, not the list

to remove the rows. You can put this inside the loop, or use lapply (to apply a function over a list), but first you need a function to be applied to every element in the list (all your 105 matrix), so

removeRows = function(m) {
ind = m[,12] == 0 | is.na(m[,12])
m   = m[!ind, ]
return(m)
}

m = lapply(m, FUN=removeRows)

That should work.