Here are a couple of REPORT step variations on your table. title1 'Value for Each Hour'; proc report data=have nowd; column ('Hour' hour),value; define hour / across ' '; define value / analysis ' '; run; proc report data=have nowd; column label ('Hour' hour) ,value; define label / computed ' '; define hour / across ' '; define value / analysis ' '; compute label / char length=6;    label='Value'; endcomp; run; ... View more

You could filter for the '.' using a WHERE.  However you will also need to use a LEFT function to eliminate leading blanks.  The period is actually being stored with two leading blanks. proc freq data=have(where=(left(z) ne '.')); ... View more

Assuming that you are using the CHAR. format shown in your post (the version mapping . to ' ') (numeric missing to character missing), you should be seeing just the columns ONE and TWO for both NUM and Z.  PROC FREQ like other procedures that work with classification variables automatically exclude observations with classification variables (NUM and Z) that take on a missing value. you could also apply a WHERE to subset the data by eliminating the missing values in an earlier step. ... View more

These are good points and you are right, the use of the trailing dot is more about style.  Either way as programmers learn to use the macro language they will ultimately be best served by understanding the tokenization process.  BTW I like your comment on the wiki article, it makes your point very nicely - thank you for the contribution. ... View more

This was one of the topics discussed during the SAS Global Forum presentation on macro lanugage best practices.  Don Henderson and I are in the process of creating a wiki article on the topic.  Please join the conversation at: http://www.sascommunity.org/wiki/Macro_Programming_Best_Practices:_Styles,_Guidelines_and_Conventions_Including_the_Rationale_Behind_Them#Always_suffixing_a_macro_variable_reference_with_a_period.2Fdot_.28..29 The consensus of those at the conference was that the dot should only be used when needed.. ... View more

As Ksharp suggests I believe that you are confusing numeric and character missing values; The numeric variable is not changed by the PUT function.  The new variable Z is character and the period (.) is no longer a missing value. Is the following something like you are seeing (you did not show much code so I am guessing)?  Notice that the missing value for NUM is automatically excluded by FREQ (it is a classification variable), while Z has no missing values. proc format; value char 1="ONE" 2="TWO"; run; data have; input num; z=put(num,char.); datalines; 1 2 . run; proc freq data=have; table num*z; format num char.; run; ... View more

To conditionally change text attributes use the CALL DEFINE routine.  Replace Ksharp's REPORT step with the following. proc format; value misswd    . = ' '    other=[best12.]; run; proc report data=x nowd; column group Demographics  count; define group/group noprint; define demographics / order order=data; define count / analysis format=misswd.; compute count; if count.sum=. then call define(_row_,'style','style={font_weight=bold font_size=14}'); endcomp; compute after group; line ' '; endcomp; run; ... View more

Here is a start.  The key is to have a dummy variable (TYPE - which for clarity here I have added in a separate DATA step) that can be used with a compute block. data have; input Demographics $  count; datalines; Age                 . Mean                25   Median              24   Range               35   BMI                 .     Mean                780  Median              777  Range               450  Gender               . Female              5    Male                7    run; data have2;    set have;    if demographics in('Age','BMI','Gender') then type+1;    run; proc report data=have2 nowd;    column type demographics count;    define type / order noprint;    define demographics / display;    define count / display 'Value';    compute after type;       line ' ';    endcomp;    run; Vertically concatenated tables such as this one pose a challenge in REPORT, and usually require some preprocessing such as was done here.  There are some examples of variations in the PROC REPORT book. ... View more

You can also often split lines by repeating the justification. as in j=l xxxxxxxxxxx j=l.  the second will split the line. ... View more

If you can go with the newer SG procedures then follow Dan's suggestions, but take a look at this paper as well. http://www.lexjansen.com/wuss/2012/142.pdf However if as your question suggests, you are not using the the SG procedures, then take a look at this paper. http://www2.sas.com/proceedings/sugi29/086-29.pdf Although these two papers are directed at GPLOT and SGPLOT the concept with the legends, symbols, and patterns are similar. If these do not clear up things for you come back and reframe your question.  There are also Macro issues in your code that may then need to be addressed. ... View more

1)     The CONTINUE transfers control to the bottom of the loop.  This is the END statement, consequently the three assignment statements are not executed when VAR is missing.  Note that since you are using a DOW loop, the DATA statement will only be executed once for each unique ID. 2) During the execution phase, non-retained variables are set to missing, when the DATA statement is executed.  In this case these will be any variables that are not on the incoming data set (A). 3) the KEEP= data set option will be applied to the data set it is associated with.  To control incoming variables use it on the data set on the SET statement.  When used with the data set(s) on the DATA statement, the KEEP= applies to the new data set. ... View more