If you have SAS Data Quality, PROC DQMATCH will do this for you. You can specify which variable(s) and match definition to use, and optionally apply multiple conditions. This generates a match code for a condition. If you do not apply a condition statement then it is a single composite condition that generates one match code. If the match codes between rows are identical, there is a degree of confidence that those entities are the same.   Here is an example using both names and addresses together.   Name Address Bill Bond 202 E Cedar St Cary NC 27511-3440 William L. Bond 202 East Cedar Street, Cary, North Carolina, 27511 Prais Hilton 100 SAS Campus Dr, Cary, NC 27513 Paris Hilton 100 SAS campus Drive Cary NC, 27513   In our data, we know that rows 1 and 2 are the same person, and 3 and 4 are the same person. Row 3 has a mispelled name, while Row 4 has the correct name. The addresses also differ in how they've been input, and one includes a ZIP+4. We need to programmatically determine who is who.   We can use the Name (with Suggestions) match definition to find similar names with mispellings, and the Address match definition for the addresses. If the combination of those two generate the same match code, we will assign them the same group.   You may need to adjust the sensitivity (between 50-100) to get the desired results. Lower sensitivity increases fuzziness, while higher sensitivity decreases fuzziness. You can mix sensitivity on a per-variable basis.     data customers; length Name Address $50.; infile datalines dlm='|'; input Name$ Address$; datalines; Bill Bond|202 E Cedar St Cary NC 27511-3440 William L. Bond|202 East Cedar Street, Cary, North Carolina, 27511 Prais Hilton|100 SAS Campus Dr, Cary, NC 27513 Paris Hilton|100 SAS campus Drive Cary NC, 27513 ; run; /* Load the USA English locale */ %dqload(dqlocale=(enusa)); proc dqmatch data = customers out = customers2 matchcode = match_cd /* Output variable name of match code */ cluster = group_id; /* Output variable name of a cluster of same match codes */ criteria var=Name matchdef='Name (with Suggestions)' sensitivity=50; /* Name must match */ criteria var=Address matchdef='Address' sensitivity=60; /* and Address must match */ run;   GROUP_ID Name Address MATCH_CD 1 Bill Bond 202 E Cedar St Cary NC 27511-3440 M&P$$$$$$$$$$$$$$$$$$!$$$$$$$$$$$$$$H00$$$$$$$$$$$4&8&Y$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ 1 William L. Bond 202 East Cedar Street, Cary, North Carolina, 27511 M&P$$$$$$$$$$$$$$$$$$!$$$$$$$$$$$$$$H00$$$$$$$$$$$4&8&Y$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ 2 Prais Hilton 100 SAS Campus Dr, Cary, NC 27513 2&W$$$$$$$$$$$$$$$$$$!$$$$$$$$$$$$$$I00$$$$$$$$$$$4&43&$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ 2 Paris Hilton 100 SAS campus Drive Cary NC, 27513 2&W$$$$$$$$$$$$$$$$$$!$$$$$$$$$$$$$$I00$$$$$$$$$$$4&43&$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$       ... View more

Here is a macro function and FCMP function that will achieve this, modified from a utility function I use called cquote.   %macro squote(strlist); %unquote(%bquote(')%qsysfunc(tranwrd(%qsysfunc(compbl(%superq(strlist))),%bquote( ),%bquote(' ')))%bquote(')) %mend; proc fcmp outlib=work.funcs.str; function squote(str$) $32767; return (cats("'", tranwrd(compbl(str),' ',"' '"), "'")); endfunc; run; Example: options cmplib=work.funcs; %put %squote(a b c); %put %sysfunc(squote(a b c)); 'a' 'b' 'c' 'a' 'b' 'c'   ... View more

Hi @Zorac, can you: Post a screenshot of what the postal code areas look like in Visual Analytics Attach the shapefile or make it available for download if it is public data When lines do not connect as expected, this is usually indicative of Visual Analytics not knowing the correct row order. In a nutshell, Visual Analytics connects the dots of vertices, and if the order of the dataset is wrong, it can look pretty weird. Data loaded to CAS is not guaranteed to have an order since it is a multithreaded, distributed environment. In order to ensure the right order, a Sequence variable is needed to tell Visual Analytics what order the shapes should be drawn.   When you load your shapefile with PROC MAPIMPORT, load it onto compute first and use PROC GREDUCE if needed. Afterwards, create a sequence variable:   data shapefile; set shapefile; sequence+1; run; Perform any additional data manipulation as needed, then load it to CAS. Specify that variable for the Sequence role when adding it as a geographic provider. In later versions of Viya, you can also do this with the ADDROWID option:   data caslib.shapefile(addrowid=yes); set shapefile; run;   ... View more

Hey @softgalaxy! Your target is interval, so you should not be running a classification model. Try fitting a model that supports interval targets instead, such as a regression, decision tree, or gradient boosting. A more appropriate target for logistic regression would be something like DeathCause or Status. ... View more

Hey @Xyp! I think this will get you what you need if I understand your question correctly. You can use PROC FREQ to do this by using the underlying table that is created for each character column. Let's say we have data like this: data have; array category[10] $; do i = 1 to 10; do j = 1 to dim(category); select(mod(i, 3)); when(1) category[j] = 'a'; when(2) category[j] = 'b'; otherwise category[j] = 'c'; end; if(i < j) then category[j] = ''; end; output; end; drop i j; run; category1 category2 category3 ... category10 a b b c c c ... ... ... ... a a a a We're expecting  a bar chart with 10 non-missings for values for category1, 9 non-missings for category2, etc.   If we run PROC FREQ, we'll see that it knows how many values there are for each level of each category, as well as the total missings: proc freq data=have; tables _CHARACTER_; run;   We want the underlying table that this output is based off of, so we're going to snag it with ODS OUTPUT. Note that if you wanted to figure out the name of this table, you can turn ods trace on to get it. ods select none; proc freq data=have; tables _CHARACTER_; ods output OneWayFreqs=OneWayFreqs(keep=table frequency); run; ods select all;   Now we have a table that looks like this: Table Frequency Table category1 4 Table category1 3 Table category1 3 Table category2 3 Table category2 3 Table category2 3 Table category3 3 Table category3 2 Table category3 3   These are our non-missing counts for each category. Let's clean up the name and plot it. sgplot will do all of the summing for us, so we don't need to do any additional post-processing. data want; set OneWayFreqs; table = scan(table, -1); rename table = category; run; proc sgplot data=want; vbar category / freq=frequency categoryorder=respdesc; run;   Hope this helps! ... View more

Great idea for making the ID a character to make the solution very elegant. Nicely done! ... View more

Hey @Li2024! PROC EXPAND can do this very quickly. It's a great swiss-army knife of multi-row operations.   proc expand data=a out=want(where=(time > 0)); by id; convert _NUMERIC_ / method=none transform=(dif 1); run; ID TIME A B C D 1 1 1 1 1 1 2 1 1 3 -1 3 3 1 2 1 -1 -1 You can ignore the TIME variable and drop it later if needed.   With a DATA Step, it looks like this: data want; set a; by id; array numvar[*] _NUMERIC_; array difvar[5] _TEMPORARY_; do i = 2 to dim(difvar); difvar[i] = dif(numvar[i]); if(NOT first.id) then numvar[i] = difvar[i]; end; if(NOT first.id) then output; drop i; run; ID A B C D 1 1 1 1 1 2 1 3 -1 3 3 2 1 -1 -1   ... View more

Quick follow-up question: When you say SAS 9.4, do you mean the SAS executable (i.e. SAS Enhanced Editor) like this?     Or Enterprise Guide?           ... View more

Hey @bussql! What is your coding environment and version? SAS Studio, SAS Enterprise Guide, or the SAS VS Code Extension? ... View more

Unfortunately you cannot use SAS input dataset options like in PROC SQL. FEDSQL follows 1999 ANSI SQL standards. FEDSQL does support the stimer option but you cannot have it print out information as it goes like you could do for a DATA Step FEDSQL does not have hints, but PROC SQL kind of does with the magic= option; however FEDSQL does have a ton of table options you can apply. Here's how to apply them.   ... View more

You can also try using FedSQL. FedSQL is multi-threaded where possible. proc fedsql; create table saslibrary.outputdataset as select t.bene_id, t.clm_id, <26 other variables> from saslibrary.lineinputdataset as t inner join saslibrary.headerinputdataset as c on t.clm_id = c.clm_id ; quit;   ... View more

Hey @acordes! I've got a macro and a function called cquote that was greatly inspired by @Tom's work and it has served me well over the years: https://github.com/stu-code/sas/blob/master/utility-macros/cquote.sas   %macro cquote(strlist, quote); %if(%upcase(&quote) = SINGLE) %then %let q = %str(%'); %else %let q = %str(%"); %unquote(%bquote(&q)%qsysfunc(tranwrd(%qsysfunc(compbl(%superq(strlist))),%bquote( ),%bquote(&q,&q)))%bquote(&q)) %mend; proc fcmp outlib=work.funcs.str; /* Double quote version */ function cquote(str$) $200; return (cats('"', tranwrd(compbl(str),' ','","'), '"')); endfunc; /* Single quote version */ function scquote(str$) $200; return (cats("'", tranwrd(compbl(str),' ',"','"), "'")); endfunc; run; You can give these a try as well. ... View more