0
$\begingroup$

I am attempting to use the monomvn package to impute monotone missings. I've followed the basic example:

data(cement.miss)
out <- monomvn(cement.miss)

However none of the outputs seem like the original input with missing imputed. I'm sure I'm missing something. Apologies in advance if this question is amateurish. I'm a computer scientist tasked with doing some statistical work =)

$\endgroup$

1 Answer 1

2
$\begingroup$

Full disclosure: I haven't used monomvn before, but glancing at the help files, it doesn't appear that it's really meant to directly impute missing data. Rather it seems intended be a way to estimate the multivariate normal distribution of data with lots of missing values, which is an important difference.

If your goal is data imputation, perhaps you should investigate some other imputation method. I'm most familiar with imputation using random forests, but there's an enormous amount of literature out there on this topic.

Maybe one way to get started would be to try running findFn("impute") from the sos package.

$\endgroup$
1
  • $\begingroup$ thanks so much! also, thanks for the pointer to the sos package $\endgroup$
    – Suraj
    Aug 11, 2011 at 13:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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