Hi:
In fact, this kind of "banding" on the headers is probably more easier to achieve with PROC TABULATE (using style=<parent>) than with PROC REPORT (which does not have style=<parent> ability). For example, see these Tech Support notes:
http://support.sas.com/kb/25/401.html (horizontal banding)
http://www2.sas.com/proceedings/sugi29/085-29.pdf (horizontal and vertical banding)
Note that the banding shown in the above links could just be applied to the headers, and not to the data cells. You still did not say where the "numbers" were coming from -- have you precalculated the count? What procedure did you use? I am not sure about the 1. and 2. in the cell for #READMITS EXPECTED, however, generally, both TABULATE and REPORT will only put 1 number in 1 cell (you show 2 "lines" in one cell). Although PROC REPORT could do something like what you show, if you built a single variable from both values and inserted a "line feed" or "carriage return" between the two values.
Without an idea of what your data looks like, it's hard to provide better advice. Instead, I am attaching a simple example with PROC REPORT (#1a and #1b) PROC TABULATE (#2) getting the counts for sales and inventory for 3 regions in SASHELP.SHOES. Not a very exciting report, number-wise, but it shows the header behavior differences (see screenshot)
cynthia
proc format;
value $regcolr 'Asia'='pink'
'Canada'='yellow'
'Pacific'='lightgreen';
run;
ods listing close;
ods html file='c:\temp\shoes_across.html' style=sasweb;
proc report data=sashelp.shoes nowd;
title '1a) Using REPORT and ACROSS';
where region in ('Asia', 'Pacific', 'Canada')
and product in ('Boot', 'Slipper');
column product region,(sales inventory);
define product / group style(column)=Header;
define region / across ' '
style(header)={background=$regcolr. color=black};
define sales / n f=comma10. ;
define inventory / n f=comma10. ;
run;
proc report data=sashelp.shoes nowd;
title '1b) Using REPORT and ACROSS and CALL DEFINE';
title2 'CALL DEFINE only changes "data cells" not header cells';
where region in ('Asia', 'Pacific', 'Canada')
and product in ('Boot', 'Slipper');
column product region,(sales inventory);
define product / group style(column)=Header;
define region / across ' '
style(header)={background=$regcolr. color=black};
define sales / n f=comma10.;
define inventory / n f=comma10. ;
compute sales;
call define('_c2_', 'style', 'style={background=pink}');
call define('_c4_', 'style', 'style={background=yellow}');
call define('_c6_', 'style', 'style={background=lightgreen}');
endcomp;
compute inventory;
call define('_c3_', 'style', 'style={background=pink}');
call define('_c5_', 'style', 'style={background=yellow}');
call define('_c7_', 'style', 'style={background=lightgreen}');
endcomp;
run;
proc tabulate data=sashelp.shoes f=comma10.;
title '2) Using TABULATE';
where region in ('Asia', 'Pacific', 'Canada')
and product in ('Boot', 'Slipper');
class product region;
classlev region /style={background=$regcolr. color=black};
var sales inventory / style=<parent>;
table product,
region=' '*(sales inventory)*n=' ';
run;
ods html close;
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.
-
2745 views
-
0 likes
-
-
