Practical Predictive Analytics

Listwise deletion

For MCAR, eliminating rows that have missing values is acceptable. In this example (from https://en.wikipedia.org/wiki/Listwise_deletion), observations 3, 4, and 8 will be removed from the data prior to analysis:

However, the downside effect of listwise deletion is that you may end up deleting some non-missing variables that could have an important effect on the model. However, since the missing data is presumed random, removing some rows would not have a major effect, assuming your assumptions are correct.

Another consideration is the amount of data you have to work with. You do not want to end up deleting most of your ...