Thank you so much for your patience!! Volume is the variable I'm trying to generate. If a data entry has the year "2010", then the code I'm looking for in OP would ideally use the variable "pct_2010" to generate a 1, 2, 3 or 4 based on <25, 25-75, >75, or missing. As for var, I'll probably end up averaging it, or doing other basic descriptive statistics. Var would be things like age, length of stays, etc, things that I want to calculate given an ID's volume. My overall research question is to find out whether "var" (such as age, length of stays) vary by not only ID, but by volume, as defined by the ID's percentile that year. I had previously calculated each ID's "volume" (e.g., volume of hospital visits) by year and attached that number to each ID, e.g. pct_2010 means the percentile of volume (of hospital visits) that that ID had relative to all other IDs that year, 2010. Thank you so much for helping me get to exactly where I need to be!! ><
... View more