Business Intelligence

To create interesting 3D Graphics which are easy to understand you can use PROC GCOUNTOUR.
With this procedure you are able to create graphics for your next presentation or for playful marketers.

The following Code and the picture are an example how to create Contour Plots:

data surface;
do i=1 to 50 by 1;
x=i;
do k=1 to 50 by 1;
x=x;
y=k;
z=round(sqrt(5*y*(x**2)+3),1);
output;
end;
end;
run;

GOPTIONS xpixels=800 ypixels=600;
GOPTIONS CBACK=WHITE;

LEGEND1
LABEL=(FONT=‚Arial‘ HEIGHT=8pt JUSTIFY=CENTER ‚Revenue:‘ )
POSITION=(MIDDLE BOTTOM OUTSIDE)
across=4
down=1
value=(tick=1 height=5.5pt JUSTIFY=LEFT ‚Low‘
tick=2 height=5.5pt JUSTIFY=LEFT ‚Low – Medium‘
tick=3 height=5.5pt JUSTIFY=LEFT ‚Medium – High‘
tick=4 height=5.5pt JUSTIFY=LEFT ‚High‘)
CSHADOW=GRAY
CFRAME=CXE8E8E8
CBORDER=GRAY;

TITLE ‚Contour Plot‘;
TITLE2 ‚PLOT: x * y * sqrt(5*y*(x**2)+3)‘;

PROC GCONTOUR DATA = surface;
PLOT x * y = z /
PATTERN
LEGEND=LEGEND1
NLEVELS=4
SMOOTH
CFRAME=WHITE
CLEVELS=GREEN YELLOW ORANGE RED
;
RUN;
QUIT;
TITLE; TITLE2; FOOTNOTE;     

If you want to import data into SAS with SAS Code you can use the following structure:

data my_table;
infile „[Type in the full Path where to find the data]“
delimiter=‚,‘
missover
dsd
lrecl=32767
firstobs=2
;
length
first_field $ 24
;
input
first_field $
;
label
first_field = „first_field“
;
run;

If you want to export data you can use the following structure. If you need more detailed code you can use the procedure output in the LOG window.

proc export data=my_table
outfile=„[Type in the full Path where to find the data]“
dbms=dlm;
delimiter=‚|‘;
run;    

With this SAS Program you are able to connect to a database with SQL Pass Through via ODBC. The advantage is, that the SQL Statement will be processed by the resources of the database. The disadvantage is, that you cannot use SAS specific functions which are not supported by the database sql language.
—
proc sql;
connect to odbc (dsn=“…“ user=“…“ password=“*****“);
create table my_table as
select * from connection to odbc
(
select
…
from
…
where
…
);
disconnect from odbc;
quit;