Dear Community,    I have the following dataset, that describes N persons who have up to 3 visits and took 3 measures (vars) at each visit.    Person     Visit   Var1 Var2 Var3    1               1      10    30     1     1               2       5     20     3  2               1       7     15     5      2               2       5      13    5     3               3       5      13     5        3               1       6      12    4          3               2       1      10    7      3               3       2       8    6         I would like to calculate a difference from the first visit for each Var. The resulting dataset should look like this:    Person     Visit   Var1 Var2 Var3    Diff_1     Diff2    Diff3 1               1      10    30     1          .            .           . 1               2       5     20     3 ..        -5          -10       2 2               1       7     15     5          .           .           .  2               2       5      13    5          -2         -17       0 3               3       5      13     5         .             .          . 3               1       6      12    4         1            -1         -1 3               2       1      10    7         -4          -3          2 3               3       2       8    6          -3          -5          1   I created the following code:  ----------------------------------------------------------------- PROC SORT DATA = have;       by Person VISIT; RUN;   DATA want;     SET have;     by Person VISIT;     RETAIN BL; //Baseline value = value for the first visit ARRAY ITEM{3} VAR1 - VAR3; ARRAY ITEM_sinceBL{3} Var1_sinceBL Var2_sinceBL Var3_sinceBL; DO I=1 TO 3;     if first.Person THEN BL = ITEM(I);     ELSE ITEM_sinceBL(I) = ITEM(I) - BL; END; drop BL; RUN; ––––––––––––––––––––––––––––––––––––––––––   (I use look because in reality I have 18 variables.) But I don't get the right result. My BL seems to have the wrong value.    I would appreciate your help very much   Julia ... View more

Dear SAS users,    I have the following "vertical" dataset that represents doctor visits for 4 patients (every patient may have one or more visits):    PatientID   visitN  visitDate              labResult 1               0            03/15/2016           10 1               1            03/15/2017            20 1               2            09/20/2017            20 2               0            06/15/2016            30  2               1            10/15/2016            30 3               0            03/15/2016            10 4               0            05/15/2016            20 4               1            08/20/2016            20  4               2            03/15/2017            20   I would like to create several new variables:   1. T1 =  to the FIRST visit (i.e., visitN = 0) . Basically, my T1 for each patient will be equal to bolded dates. 2. nDaysSince1stvisit - number of days since T1 (i.e., since the first visit)   May somebody direct me to how to do it?   Thank you very much ahead.  Julia   P.S. Eventually, I would like to get a "horizontal" dataset with 1 line per patient and with the labResults data for T1 (first visit), T2_1month (the visit in 1 month since the first one), and T3_1year (the visit in 1 year since the first one). However, to avoid confusion, I thought I would rather ask step by step.  ... View more

Dear all,    I am using the following code to create a 'nice" output of frequencies and % for several variables:    PROC TABULATE DATA=PATH.survey_formatted MISSING ORDER=FORMATTED; TITLE6; TITLE7 J=CENTER HEIGHT=2 COLOR=grey 'Table 1. Demographic Characteristics of Participants'; CLASS sex agecat racecat location education_level ; CLASSLEV sex agecat racecat location education_level  / STYLE=[cellwidth=3in asis=on]; TABLES (sex agecat racecat location education_level)* (n*F=4.0 pctn='(%)'*F=pctfmt.)/MISSTEXT='0' RTS=10;   The output, however, looks like one long horizontal table with 1 row and multiple columns. I would like it to look more like a table with many rows and 3 columns: variable name, Freq, % :    ---------| N     |    %   | sex       |         |          |    M      | 20    |  10% |    F       | 180  | 90%  | ------------------------- race       |    White  |    AA      |    Asian  |    Other  |    How can it be done? Thank you very much.    Julia ... View more

Thank everybody for the previous help.  I have one more question.    I am trying to merge two datasets:    data dataSet1;  type1 something1  type1 something2  type1 something3   type2 something1  type2 something2  type2 something3   type3 something1  type3 something2  type3 something3    data types;   input myText textType;   datalines;    a  type1     b  type2     c type3 run;   I would like to merge these two to get    type1 something1 a  type1 something2 a  type1 something3 a   type2 something1 b  type2 something2 b  type2 something3 b   type3 something1 c  type3 something2 c  type3 something3 c   After this I can get rid of Type variable completely. So, basically, I would like to replace type1 by the corresponding 'a', type2 by the corresponding 'b' etc.   The simple code:     data dataset1;   merge dataset1 types;   BY textType; RUN;     creates a strange result:     type1 something1 a  type1 something2   type1 something3    type2 something1 b  type2 something2   type2 something3    type3 something1 c  type3 something2   type3 something3    (so, textType appears only  first time and all the rest are missing values. What is wrong here? What is the most efficient way to do it?     Thank you very much   Julia ... View more

Dear Sas experts,    I am trying to create a simple dataset with 2 char variables, one of each is actually a long string with unpredicted length.    data myLine; infile datalines delimiter='|'; input myLine $75. lineType $20.; datalines; L1. This is my very long line of unpredicted length. | line_one L2. This is my second very long line of unpredicted length.| line_simple L3. This is my third, third, third very long line of unpredicted length.| line_another ; run;   However, the dataset looks weird, e.g.    myLine                                                                              lineType This is my very long line of unpredicted length. | line         one   The results really depends on the length of the first column (myLine). The numbers .75 and .20 are arbitrary. When I put no numbers (just $) the lines are cut after 8 symbols, I believe.    What is the right way to allow unlimited length (or at least limited by 200 symbols) for the line and to create reliable column values.    Thank you ... View more