Use DatePART() function to convert datetime to date and subtract the two. ... View more

You didn't provide any raw data, so I create a dummy data set that has an age variable (random ages) and a group variable with values of 100, 200 and 300. data have;    do group= 100, 200, 300;       do i= 1 to group;          age = round( 100*ranuni(345),1);          output;       end;    end; run; /* a custom format. Your example says that you want a multilabel since values >75 are also >65. Only a few procedures will actually honor multiple label type variables */ proc format library=work; value myage  (multilabel) low - < 65 = '< 65 years' 65 - high  = '>= 65 years' 75 - high  = '>= 75 years' ; run; /* This is one of the procedures that will use the multilabel format indicated by MLF on the class statement this come close to demonstrating the output you showed. We can modify the format to round the row percentage display. If you really need () then there is a bit more work but that is for a separate thread after getting the basics*/ proc tabulate data=have;    class age / mlf;    format age myage.;    class group;    tables age, group=''*(n='' rowpctn=''); run; ... View more

These and several other random number functions return a random value from a distribution. Ranuni from a uniform distribution on the interval from 0 to 1 and rannor from a normal distribution with a mean of 0 and standard deviation of 1. Uses are many depending on your needs but basically revolve around a need for a random number. One example would be to add a value to set of data and use a specified subset of the data to select a random sample. For instance add a value x=ranuni (12345); to each record of a data set and then select x < .20 to get roughly a 20 percent random sample of the data. ... View more

Problem with automating extractions. I was initially going to say contact tech support, but then realized that the question/answer wasn't in a standard SAS format and figured it came from a site like this.  At any rate, the answer above is wrong. You get what you pay for in these cases :smileymischief: ... View more

Using the format as part of an input statement occasionally has issues with column alignment if you aren't providing fixed columns or delimited data to read from. Standard dates aren't much of an issue though custom formats and string values can be entertaining on occasion. Also if you have an INFORMAT supplied on the INPUT it overrides the INFORMAT statement. The column order of the variables can be set with an informat or length statement(s) prior to input if you would like them to appear in other than supplied data order as well. See the result here: data aes; label USUBJID  = 'Unique Subject Identifier'            AESTDTC = 'Start Date/Time of Adverse Event'            AEENDTC = 'End Date/Time of Adverse Event'            AETERM = 'Reported Term for the Adverse Event'; informat AESTDTC  date9. AEENDTC  date9.; input @1 USUBJID $3. AETERM $15. AESTDTC date9. AEENDTC date9. ; datalines; 101  Headache 01JAN2004 02JAN2004 101  Back Pain 15JAN2004 03FEB2004 102  Rash 03NOV2003 10DEC2003 102  Abdominal Pain 03JAN2004 10JAN2004 102  Constipation 04APR2004 04APR2004 ; run; Compared with: data aes2; label USUBJID  = 'Unique Subject Identifier'            AESTDTC = 'Start Date/Time of Adverse Event'            AEENDTC = 'End Date/Time of Adverse Event'            AETERM = 'Reported Term for the Adverse Event'; informat AESTDTC  date9. AEENDTC  date9.  AETERM $15.; input @1 USUBJID $3. AETERM AESTDTC AEENDTC ; datalines; 101  Headache 01JAN2004 02JAN2004 101  Back Pain 15JAN2004 03FEB2004 102  Rash 03NOV2003 10DEC2003 102  Abdominal Pain 03JAN2004 10JAN2004 102  Constipation 04APR2004 04APR2004 ; run; It also affects how spaces in data may be interpretted ... View more

Your data contains TAB characters on some lines. Try adding the statement INFILE DATALINES EXPANDTABS; to your datastep. - PG ... View more

Hi:   This is a question for SAS Tech Support. If you look inside the BAT file contained in the zip archive using Notepad, it appears that this bat file is intending to set a Main SAS directory variable in a location called C:\Program Files\SAS -- since you do NOT have a C:\Program Files directory for SAS with SAS University Edition (since it is inside the Virtual Machine) and since you cannot do any installations inside the Virtual Machine, my guess is that you are not going to be able to do this install or use the Toolkit. cynthia ... View more

Thanks!Now I  understood clearly.... ... View more

Just to reiterate, you can create permanent libraries, place the libname code in the autoexec or use the GUI menu. It was probably created by modifying the auto exec somehow. AFAIK it  is a limitation in all SAS BASE/UE versions. ... View more