2
\$\begingroup\$

I have this for loop for creating a list of variables:

vars <- c( 'beta.o', 'sd.y') 
for (x in c('ghs', 'site', 'trt')) {
  if(model.parms[[x]] == 1) {
    data <- data[, which(names(data) != x)]
  } else {
    data <- data
    if(x!='ghs') {
      vars <- c(vars, paste('sd.', x, sep = ''))
    }
    m <- min(model.parms[[x]], 5)
    for (i in 1:m) {
      if(i == 1 && x == 'site') {
        vars <- c(vars, 'beta.site[1]')
      }
      if (i > 1) {
        vars <- c(vars, paste('beta.', x, '[', i, ']', sep=''))
      }
    }
  }
}

It has been bothering me terribly, and I have failed the last two times I have tried to replace it, although conceptually it should be able to be written in a few lines. Any advice?

\$\endgroup\$
3
\$\begingroup\$

This bit:

for (i in 1:m) {
  if(i == 1 && x == 'site') {
    vars <- c(vars, 'beta.site[1]')
  }
  if (i > 1) {
    vars <- c(vars, paste('beta.', x, '[', i, ']', sep=''))
  }
}

Says handle the first pass differently from all the others. So I would replace it with this:

if (x == 'site') {
  vars <- c(vars, 'beta.site[1]')
}
for (i in 2:m) {
  vars <- c(vars, paste('beta.', x, '[', i, ']', sep=''))
}
\$\endgroup\$
1
\$\begingroup\$

The big mistake in your code is you are dynamically growing your vectors. For example, compare

x = NULL 
for(i in 1:100000)
   x = c(x, i)

with

x = numeric(100000)
for(i in 1:100000) 
   x[i] = i   

The other key point is that paste function can be vectorised. So,

for (i in 1:m) {
  if(i == 1 && x == 'site') {
    vars <- c(vars, 'beta.site[1]')
   }
   if (i > 1) {
     vars <- c(vars, paste('beta.', x, '[', i, ']', sep=''))
   }
 }

can be replaced with

if(x == 'site') {
    vars <- c(vars, 'beta.site[1]')
   }
vars = c(vars, paste('beta.', x, '[', i, ']', sep=''))
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.