Thanks Scott, It seems the data import method I was using  "libname  xlsxfile "C:\Users\Z\Documents\DATA\file.xlsx";"  followed by a data AGS; set xlsxfile."Sheet1$"n(dbSASType=(Time_Stamp=datetime )); run; run;was messing things up somehow. I haven't figured out how yet, but I'm guessing I occasionally get rounding error such that a point ends up in the next batch. Anyway, it wasn't an issue when I cut and pasted the data into SAS. Richard - I tried the SQL solution, but would receive an error message (22-32 expecting a format name) for the format = datetime line. I'm not sure if this is also associated with my import method. ... View more

Scott, Thanks - this is what I was looking for but it is not quite working and I can't figure out why (unfortunately I have no experience with either hash objects or proc SQL). For the first hour of data, this code ends up averaging the following "bins" of data: minutes 01-10, 11-19, 20-29, 30-39, 40-49, 50-59. The next hour is then 00-09, 10-20, 21-30, 31-40, 41-49, 50-59. So, in the first hour the 11-19 segment is averaging across 9 min instead of 10 and in the second hour the 10-20 segment is averaging across 11 min instead of 10 with the 41-49 segment averaging across 9 minutes again. Any Ideas on why this might be happening? Thanks. ... View more