Hi ... I had a similar idea ... why not just use SUMMARY then a data step.  One issue that arose is the MONTH is a literal rather than a number and a CLASS statement puts the months in alpha order rather than chronological order (you'd have to muck around with the data add month as 1, 2, 3 etc) .... proc summary data=x; class agency program month; types agency*month program; var active_cases pending_cases housing_subsidy_cases program_capacity; output out=stats (drop=_:) sum=; run; data y; set stats; utilization=active_cases+pending_cases+(housing_subsidy_cases / 2) / program_capacity; run; agency     program     month         cases       cases       cases     capacity    utilization            ProgramA                   810         174         84         1120        984.038            ProgramB                   405          87         42          560        492.038 Agency1                April           26           7          4           40         33.050 Agency1                August          37           9          4           50         46.040 Agency1                December        44          10          4           60         54.033 Agency1                Febuary         27           6          4           40         33.050 Agency1                January         26           6          3           40         32.038 Agency1                July            38           5          4           50         43.040 Agency1                June            32          10          4           40         42.050 Agency1                March           28           8          4           40         36.050 Agency1                May             30           6          4           40         36.050 Agency1                November        42           8          3           60         50.025 Agency1                October         37           5          1           50         42.010 Agency1                September       38           7          3           50         45.030 Agency2                April           52          14          8           80         66.050 Agency2                August          74          18          8          100         92.040 Agency2                December        88          20          8          120        108.033 Agency2                Febuary         54          12          8           80         66.050 Agency2                January         52          12          6           80         64.038 Agency2                July            76          10          8          100         86.040 Agency2                June            64          20          8           80         84.050 Agency2                March           56          16          8           80         72.050 Agency2                May             60          12          8           80         72.050 Agency2                November        84          16          6          120        100.025 Agency2                October         74          10          2          100         84.010 Agency2                September       76          14          6          100         90.030 ... View more

data decode; q=1;n=2; d=2; output; q=2;n=0; d=2; output; q=3;n=0; d=0; output; q=4;n=1; d=2; output; q=.;n=0; d=0; output; run; %macro decode; data want; merge  %do i=1 %to &n;          decode(rename=(q=q&i n=q&i.n d=q&i.d))         %end;        ; run; %mend; %let n=4 ; %decode Ksharp ... View more

hi ... removing formats is one of the tasks that you can also accomplish with PROC DATASETS rather than a data step if the data set is large, you might save some time with ... proc datasets lib=<whatever> nolist; modify want; format _character_; quit; ... View more

It depends on the relation between your disposition date and the last date of the month. You'll need to provide some more information on your criteria and some sample data with expected output is always helpful. If you had a fixed date something like the following would work. Proc sql; create table want as select distinct var1,  var2, var3 from have where datepart(disposition_date) <= '31jul2012'd ; quit; ... View more

Hi:   I don't quite understand what you want, then. Your PROC REPORT code already shows one of the methods to change the style of the spanning header -- your header should be the same style (background, color, font, etc) as the other headers because of this line in your code (blue😞    proc report data=All_Spec_Med_Qs_TotalScore nowd headline headskip        style(column)={just=c font_size=11pt font_face= 'Arial Narrow'cellwidth=2in}        style(header)={font_size=11 pt font_face= 'Arial Narrow' FONT_WEIGHT =BOLD background=cxcccccc};        column  ('Special Medical Review Overall Score'  totalAwarded  Total_Possible  Final_score) ;   That style override should also have an impact on your spanning header.  Are you saying that you do NOT want the spanning header to be the same style as the other headers? How would you change the string 'Special Medical Review Overall Score' to be different??? You could just put that string into a TITLE statement or an ODS TEXT= statement. Usually when you put a spanning header into your report table, above the other column headers, the assumption is that you want the spanning header to be the same style as the other headers, this is why STYLE(HEADER) on the PROC REPORT statement has the effect that it does have. cynthia Added: It also seems like you were trying to use a TITLE statement in this previous post: https://communities.sas.com/message/127085#127085 ... is that posting related at all to this posting? Do you have 1 report or 3 reports? Are you also using STARTPAGE=NO with PDF? If so, in PDF, titles from the first procedure are the only titles that will appear on the page. So if you need some kind of sub-title between each procedure, then you are on the right track with spanning headers. Another alternative is ODS TEXT between procedures. Is your question really about spanning headers or is it about titles? ... View more

Hi Mgarret, if you want to put all report tables on one page then don't use title. Use ods pdf text=""; instead. ods pdf text="your first title"; proc report....; run; ods pdf text="your second title"; proc report...; run; etc. Best wishes Eva ... View more

There are many graph examples discussed on Graphically Speaking. ... View more

Hi:   You may need to work with Tech Support on this. The flyover popup box is interpreted/rendered differently by each destination that allows flyover. I'm not sure there is a way to impact the flyover text. If there is, my guess is that it would be a template change. I don't know whether ODS ESCAPECHAR style overrides would work, but I suspect not. cynthia ... View more

Since there are some values that overlap (e.g., -2LOW, 0-Low, and LOW) is the order you listed the order of prescedence that you want applied? ... View more

Hi:   The issue is NOT SAS, but is Excel and how Excel renders the HTML file that you create with ODS TAGSETS.MSOFFICE2K_X. No matter what cellwidth you specify, Excel might not respect the cellwidth. It has defaults for column type and column width and frequently does whatever it wants and not what you specify in SAS.   The "fix" is to use the HTMLSTYLE attribute to send a Microsoft format from SAS to Excel. I'm not on a computer with fast Internet connections, so I will just refer you to the support.sas.com web site and search for HTMLSTYLE and mso-number-format. Or, search for my SGF paper "Don't Gamble with Your Formats" -- from SGF 2011. I had an example of setting width with HTMLSTYLE in the paper. cynthia ... View more

Hi:   There have been MANY previous forum postings about the differences. The features are really up to you whether they're pros or cons. If the inclusion of images or logos is important to you than you might stick with HTML methods and ODS MSOFFICE2K. If multi-sheet workbooks is important to you and you do not have graphs or images to include, then ODS TAGSETS.EXCELXP is a better choice.   Bottom line is that neither method creates "true, binary, proprietary" Excel files. When you use ODS methods, you are creating ASCII text files that Excel knows how to open and render so that you report looks like a workbook with worksheets. If you open your files from ODS with Notepad you will see either HTML tags or XML tags in the file: ODS MSOFFICE2K creates Microsoft "flavor" of HTML tags ODS TAGSETS.EXCELXP creates Microsoft "flavor" of XML tags for Spreadsheet Markup Language 2003 cynthia Short Comparison:                       ODS MSOFFICE2K             ODS TAGSETS.EXCELXP What Is Created   Microsoft "flavor" HTML      Microsoft "flavor" of XML                   created in Office 2000       created for Spreadsheet Markup                                                  XML as defined in Office 2003        How Excel renders Uses SAS style               Uses SAS style                   Only 1 sheet in 1 workbook   Multiple sheets in 1 workbook       Include Images?     Yes                             No     How to override Excel defaults    STYLE= override (HTMLSTYLE)  STYLE= override (TAGATTR)    Version Excel     Office 2000 and higher       Office 2003 and higher ... View more

Hi,  Try this... data have; input Year $ Office  $ Safey  WellBeing  AdminSupport stafftrai; cards; FY11 Office1 .90 .67 .78 .99 ; run; proc transpose data=have out=want; var year office Safey  WellBeing  AdminSupport stafftrai; run; ... View more

Then use lists. %let final=9; data want ;    set have ;    keep Quest Program_Name      ChildAge1-ChildAge&final      Physical1-Physical&final      Education1-Education&final      Emotional1-Emotional&final      Social1-Social&final   ; run; ... View more