So you don't have a DATE (number of days).  You have a DATETIME (number of seconds).   The YEAR() function needs a DATE value.  You can convert a DATETIME value to a DATE value by using the DATEPART() function. ... View more

Example of the data before the Proc Summary, in the form of a data step that will create the correct variables. The code used in proc summary.   Do you really expect a data set? That output looks more like a report and I really consider placing anything resembling summary data, your "total lines items", inside a data set as it is too easy to use that total accidentally in other procedures. ... View more

@mauri0623 wrote: I want to be able to form a where statement based on three variables. One variable is called edipn and the others are patcat_e and patcat_p. basically I want to say, where edipn not equal to missing and the first to 5th position of the patcat_e and patcat_p. Please help I'm afraid it isn't clear to me what you want. This part is very clear: "where edipn not equal to missing", but the rest "the first to 5th position of the patcat_e and patcat_p" is not clear to me. Please explain further, provide more details. Examples of the data in EDIPN and the desired result of the WHERE statement would be helpful as well. ... View more

Thank you. I appreciate your time. Obviously, you are a very versed programmer. If it is OK I would like to contact you about this. I still have other part of this that I also need a little help with. I however do not want to be demanding. Just if you have a few minutes that you can spare. Any day or any time will be fine. Hope you have a blast weekend. Thank you again. ... View more

"identify the index encounter, which is the encounter_date closest to the start_date"   Questions: What if there is a tie?  If one date precedes start_date by the same number of days as another follows start_date, which one is selected as the index date. As per @ballardw 's note, what to you mean by "one week" and "two weeks"?  Do you mean exactly 7 or 14 days?  Or do you ignore weekends or other days? Does each person_id has only a single start_date value, as in your sample date?   If so, do all your person_id's have the same start_date, as in your sample data? What do you want the output data to look like?  Or if it is a report, what do you want it to look like? If there is an encounter in the washout period, in what way will the output be different? ... View more

@mauri0623 wrote: I tried index but the values are coming up as numbers. Of course they are. Index returns a position number of where the value is found and 0 when not found. So typically if the criteria is "somevalue somewhere in the variable" then code often looks like: If index(variable,'somevalue') > 0 then <do something>; where the "do something" could be assign a value to a variable, delete the record, start a DO/End block of code to do something more complex. ... View more

Duplicate post merged. ... View more

Duplicate post merged. ... View more

Please explain when you want to display NA instead of the numeric value?  The code in your question is confused. if ag_code = "084500" then do; _2019_1 = put(_2019,8.); _2020_1 = put(_2020,8.); _2019_1 = _2019; _2020_1 = _2020; if _2019_1 = 0 then _2019_1 = "NA"; if _2020_1 = 0 then _2020_1 = "NA"; end; You appear to be creating _2019_1 as a CHARACATER variable, but then later you are comparing if it equals the number zero (not a string with the 0 digit in it).  Did you want the IF statements to use the original variable instead? if ag_code = "084500" then do; _2019_1 = put(_2019,8.); _2020_1 = put(_2020,8.); _2019_1 = _2019; _2020_1 = _2020; if _2019 = 0 then _2019_1 = "NA"; if _2020 = 0 then _2020_1 = "NA"; end; Note you could make a custom format to display NA when the value is zero (or any other values). proc format; value na (default=8) 0='NA'; run; Now there is no need to change the actual dataset. Just change the FORMAT attached the variable so that values of zero will print as NA. proc print data=c97 ; format _2019_1 _2020_1 na.; run; ... View more

Since the results of this merge are extremely dependent on the values of the IPAC variable in both data sets we cannot tell with the data and what your expectations actually are.   What makes you think that you are "not getting the right number for match and un-match.". As in example data values, not some "the numbers aren't right" statement. Provide an example of IPAC values that appear in the matched data that shouldn't and values of IPAC that should not appear in the matched.   Some common issues generally related to the BY variable not being exactly the same value in both sets.    If IPAC is numeric then there may be a decimal portion that you don't see that has values that might appear identical that are not. The fix if this is the case is to round the variables in both sets before merging to a standard number of decimal places.   For character variables the BY is going to be case sensitive, "ABC" does not match "Abc". The fix here is make sure the case is the same in both data sets, use any of UPCASE, LOWCASE or PROPCASE functions on the variable in both sets.   Sometimes you may have a leading space(or multiples) and "  ABC" does not match "ABC". Trailing spaces won't have any impact on the comparison but leading spaces does. The fix may be as simple as using the LEFT function on both datasets. If the "space" is actually a different non-printable character this may get more complicated.   If the format for the IPAC variable is shorter than the length you may see "ABC" (format $3.) when the value is actually "ABCDEF" (length 6 or more). This would likely require either changing the values of the variable or possibly creating a separate variable in each set to have identical values to merge on.     ... View more

@mauri0623 wrote: Yeah, I wish I know what Phillip used back in 1958. How about some inverse relationship between inflation and unemployment rate based on the data that is attached. That is a clear question. Is it?   https://onlinelibrary.wiley.com/doi/full/10.1111/j.1468-0335.1958.tb00003.x ... View more