very nice and clear - thank you
a couple of requests for follow-up blogs:
1. there are other ways to deal with censoring caused by limits of detection; a comparison, including the use of multiple imputation would be very interesting
2. what if one has both truncation and censoring (left-truncation and right-censoring is actually not uncommon in medicine). it would be great if you could extend the discussion to include both form of restricted sampling including showing the actual Stata code for dealing with this in survival analysis; note that I do not mean to limit the types of co-existence to left truncation and right censoring as other forms surely exist