I am trying to produce the following table:     Females   Males     Group Test n % n % Total Group a Test 1             Test 2           Group b Test 1             Test 2           Group c Test 1             Test 2           Total Test 1             Test 2             The closest I have come is only one row for the "total", but I want to show the overall frequency for each test. The code I have now is:   PROC REPORT DATA=have ; COLUMNS groups test sex, (n pct) all; DEFINE groups / group ; DEFINE test / GROUP "test "; DEFINE sex/ ' ' across; DEFINE pct / "%" computed format=percent8.; DEFINE all / "Total" computed; /* Sum total number of responses */ compute before test ; den = _c3_ + _c5_; endcomp; /* counter to shade rows*/ compute before groups ; I+1; ENDCOMP; compute groups ; if mod(i,2) eq 1 then call define(_row_, "style", "STYLE=[background=lightgrey]"); endcomp; /* percentage */ compute pct; _c4_ = _c3_ / den; _c6_ = _c5_ / den; endcomp; compute all; all = _c3_ + _c5_; endcomp; rbreak after / summarize; compute after /style={just=r}; groups = "Total"; _c4_ = _c3_ / all; _c6_ = _c5_ / all; endcomp; RUN; ... View more

I have data with a date column, and I would like to filter only rows that are within 1 month (before and after) of a specified month - but doing this over each year. Input data is: ID Month DATE 1 1 2019/01/03 2 2 2019/02/20 3 3 2019/03/21 4 4 2019/04/25 5 5 2019/05/12 6 6 2019/06/14 7 7 2019/07/01 8 8 2019/08/03 9 9 2019/09/20 10 10 2019/10/21 11 11 2019/11/25 12 12 2019/12/12 13 1 2018/01/03 14 2 2018/02/20 15 3 2018/03/21 16 4 2018/04/25 17 5 2018/05/12 18 6 2018/06/14 19 7 2018/07/01 20 8 2018/08/03 21 9 2018/09/20 22 10 2018/10/21 23 11 2018/11/25 24 12 2018/12/12 25 1 2017/01/22 If I were to specify month=3, I would want to keep: ID Month DATE 2 2 2019/02/20 3 3 2019/03/21 4 4 2019/04/25 14 2 2018/02/20 15 3 2018/03/21 16 4 2018/04/25 And if I were to specify month=12:  ID Month DATE 1 1 2019/01/03 11 11 2019/11/25 12 12 2019/12/12 13 1 2018/01/03 23 11 2018/11/25 24 12 2018/12/12 25 1 2017/01/22 I'm not sure how to do this by only specifying the month to use. The first example I can simply say month+1 and month-1, but when it's the year change, I'm unsure.  ... View more

I have to roll-up a lot of locations but I'm able to look for keywords to classify them. I can accomplish this with multiple IF statements but would prefer to use IN and a wildcard or FINDW operator. I'm wondering if I can combine these. Here is the original method:   DATA want; SET have; if location = "STATE1A, CITY1" then new = "Group 1"; if location = "STATE1A, CITY2" then new = "Group 1"; if location = "STATE2B, CITY3" then new = "Group 1"; if location = "STATE2B, CITY4" then new = "Group 1"; if location = "STATE3A, CITY1" then new = "Group 2"; if location = "STATE3A, CITY2" then new = "Group 2"; if location = "STATE4B, CITY3" then new = "Group 2"; if location = "STATE4B, CITY4" then new = "Group 2"; RUN; You can see that I'm only concerned about STATE1, STATE2, etc, regardless of the city. I need to search each observation for the correct State name (there can be 5-6 different ones), and I'm not concerned about the city name. The state name is varying lengths in the string. If I wanted to look for just one state name I would do the below, but I need to include a few states within the IN() command. IF FINDW(lowcase(location),"state1") > 0 then new = Group a; I would like to try: IF location IN(%STATE1A%, %STATE2B%) then new = "Group 1"; IF location IN(%STATE3A%, %STATE4B%) then new = "Group 2";   I'm unsure if I should be using the pattern matching with %, or the FINDW command within the IN()?   ... View more

  I have a data set in this form:   data have; input id x1-x5; datalines; 1 3 . . . . 2 . 2 . . . 3 . . 5 . . 4 . . . 7 . 5 . . . . 7 ; I'm trying to use an array that will create a new variable to the non-missing value. I want my new variable to look like:   x_new 3 2 5 7 7   My do loop is only selecting the final value (x5), how do I get it to select the non-missing.  ... View more

I have some .sas file that is run from a .bat file as such:   @echo off # Run the SAS code start sas "C:\file\to\code\runME.sas" The SAS program creates a PDF report, and I want the user to be able to define the time period the report is for. Is there anyway I can require the command window (or anything trigger by running the .bat file), to ask for a start and end date, that is then passed to SAS to report on the date range? ... View more

I have data:       data have; input id type $ date $; datalines; 1 c 2019-08-01 1 a 2019-08-04 1 b 2019-08-05 1 c 2019-08-11 2 c 2019-10-13 2 c 2019-10-14 2 a 2019-10-20 2 b 2019-10-25 ;     I want to remove observations that occur before type a, if they are not type a. Therefore, each ID should have their first observation being equal to type a.    Desired output:     data want; input id type $ date $; datalines; 1 a 2019-08-04 1 b 2019-08-05 1 c 2019-08-11 2 a 2019-10-20 2 b 2019-10-25 ;     ... View more

I have data in this form:   data have; input start $ end $ year; datalines; 1 01APR2000 07MAR2001 2000 1 01APR2000 07MAR2001 2001 2 25JUN2000 07FEB2001 2000 2 25JUN2000 07FEB2001 2001 ; I want to add a new column that lists the number of the month, for every month in the year, starting and ending at the existing columns (start and end). I am trying to achieve (for the first person as an example):     ID    start                  end                    year         month 1     01APR2000      07MAR2001       2000        4 1     01APR2000      07MAR2001       2000        5 1     01APR2000      07MAR2001       2000        6 1     01APR2000      07MAR2001       2000        7 1     01APR2000      07MAR2001       2000        8 1     01APR2000      07MAR2001       2000        9 1     01APR2000      07MAR2001       2000        10 1     01APR2000      07MAR2001       2000        11 1     01APR2000      07MAR2001       2000        12 1     01APR2000      07MAR2001       2001        1 1     01APR2000      07MAR2001       2001        2 1     01APR2000      07MAR2001       2001        3   I've tried a do loop but I'm not sure how to begin back at 1 after reaching 12.     ... View more

I have data for people in this form:   DATA have; INPUT id start_age end_age; DATALINES; 1 27 30 2 15 15 3 30 32 ; RUN I want to add a new row for every year of data between the two ages, inclusive of start and end ages. My desired output looks like:   ID   age_year  1          27 1          28 1          29 1          30 2          15 3          30 3          31 3          32   Should I use an array and within a do loop add rows for i=start_age to end_age? ... View more

I am trying to count observations with two grouping variables, with the output include counts of zero in groups. I've look at this question but having troubling expanding it with two grouping variables.   DATA test_dat; INPUT ID group $1. varx; DATALINES; 1 a 63 1 a 67 1 b 81 1 b 63 2 b 11 2 b 63 3 a 53 3 b 83 ; RUN; I want my output to look like:   ID     group    count 1         a            2 1         b            2 2         a            0 2         b            2 3         a            1 3         b            1   ... View more