I'm curious if you've ever encountered the following scenario:
You suspect that an issue that has been highlighted in the wider econometric literature is present in your data, but you don't feel comfortable trying to address it due to technical limitations - either a lack of a feasible method or lack of certainty about the validity of the results.
Has anything like this happened to you? What was it you were looking to do and what were the technical obstacles?