While working with date variables in STATA, I encountered an interesting problem. There is a date variable and there are hundreds of other common variables. I have to find out the initial and ending date for each variables based on that single date variable. Such as, if a variable has all missing observations up to 1st December but at least 1 respondent responds on 2 December, then the initial date for that variable would be 2 December and so on. But I would like to request for clues about how this can be done for all the variables through STATA command (or any other software) and not by looking in the actual data which is gruesome for a large dataset.