Export Reviews, Discussions, Author Feedback and Meta-Reviews

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. The paper deals with the problem of identifiability in the presence of various generated mechanisms for missing data, using a framework introduced by Mohan and others. Some existing results are improved upon, and some nice examples are given (such as identifiability of causal distributions even when conditional distributions are unavailable). Quality Technical quality seems high, the results are very involved but appear to be correct. Results like Theorem 3 are clear and interpretable.