BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
braam
Quartz | Level 8

I'm wondering why my code below starts from OBS of 2, not 1.

 


%let data= sashelp.cars;
%let FN= 1;
%let LN= 50;
%let column= _all_;
%let Gvar= make;
%let disp_Gvar= DISP_&Gvar.;
%let temp_Gvar= temp_&Gvar.;
%let cnt_Gvar= ctn_&Gvar.;
%let hold_Gvar= hold_&Gvar.;

	proc report data=&data. (firstobs= &FN. obs= &LN.) ;
		column obs &disp_Gvar. &column;
		define obs 			/computed;

		define &Gvar. 		/order noprint;
		define &disp_Gvar. 	/computed;

		compute obs;
			_obs + 1;
			obs = _obs;
			endcomp;

		compute before &Gvar.;
			&cnt_Gvar. + 1;
			&hold_Gvar.= &Gvar.;
			endcomp;
		compute &Gvar.;
			length sval $100 bk $20;
			if mod(&cnt_Gvar., 2) = 0 then bk= 'cxc4efd3'; else if bk= 'cxdddddd';
			sval= catt('style={background=', bk, '}');
			call define (_row_, 'style', sval);
			endcomp;
		compute &disp_Gvar. / character length= 20;
			&disp_Gvar. = &hold_Gvar.;
			endcomp;
		run;
obs DISP_make Model Type Origin DriveTrain MSRP Invoice EngineSize Cylinders Horsepower MPG_City MPG_Highway Weight Wheelbase Length
2 Acura MDX SUV Asia All $36,945 $33,337 3.5 6 265 17 23 4451 106 189
3 Acura RSX Type S 2dr Sedan Asia Front $23,820 $21,761 2 4 200 24 31 2778 101 172
4 Acura TSX 4dr Sedan Asia Front $26,990 $24,647 2.4 4 200 22 29 3230 105 183
5 Acura TL 4dr Sedan Asia Front $33,195 $30,299 3.2 6 270 20 28 3575 108 186
6 Acura 3.5 RL 4dr Sedan Asia Front $43,755 $39,014 3.5 6 225 18 24 3880 115 197
7 Acura 3.5 RL w/Navigation 4dr Sedan Asia Front $46,100 $41,100 3.5 6 225 18 24 3893 115 197
8 Acura NSX coupe 2dr manual S Sports Asia Rear $89,765 $79,978 3.2 6 290 17 24 3153 100 174
10 Audi A4 1.8T 4dr Sedan Europe Front $25,940 $23,508 1.8 4 170 22 31 3252 104 179
11 Audi A41.8T convertible 2dr Sedan Europe Front $35,940 $32,506 1.8 4 170 23 30 3638 105 180
12 Audi A4 3.0 4dr Sedan Europe Front $31,840 $28,846 3 6 220 20 28 3462 104 179
13 Audi A4 3.0 Quattro 4dr manual Sedan Europe All $33,430 $30,366 3 6 220 17 26 3583 104 179
14 Audi A4 3.0 Quattro 4dr auto Sedan Europe All $34,480 $31,388 3 6 220 18 25 3627 104 179
15 Audi A6 3.0 4dr Sedan Europe Front $36,640 $33,129 3 6 220 20 27 3561 109 192
16 Audi A6 3.0 Quattro 4dr Sedan Europe All $39,640 $35,992 3 6 220 18 25 3880 109 192
17 Audi A4 3.0 convertible 2dr Sedan Europe Front $42,490 $38,325 3 6 220 20 27 3814 105 180
18 Audi A4 3.0 Quattro convertible 2dr Sedan Europe All $44,240 $40,075 3 6 220 18 25 4013 105 180
19 Audi A6 2.7 Turbo Quattro 4dr Sedan Europe All $42,840 $38,840 2.7 6 250 18 25 3836 109 192
20 Audi A6 4.2 Quattro 4dr Sedan Europe All $49,690 $44,936 4.2 8 300 17 24 4024 109 193
21 Audi A8 L Quattro 4dr Sedan Europe All $69,190 $64,740 4.2 8 330 17 24 4399 121 204

 

 

 

To get what I would like to have, the below change works for me.

 

*WRONG;
		compute obs;
			_obs + 1;
			obs = _obs;
			endcomp;


*RIGHT;
		compute obs;
			_obs + 1;
			obs = _obs - 1;
			endcomp;
1 ACCEPTED SOLUTION

Accepted Solutions
Cynthia_sas
SAS Super FREQ

Hi:

  The reason you are getting "odd" calculated values for your fake obs numbers is because PROC REPORT is calculating your value on EVERY row it writes and EVERY break. So you have a header row, that is row #1 and technically, the first data row on the report is absolute row #2. And, PROC REPORT is doing something at the break for MAKE so that is throwing off your COMPUTE block.

 

  PROC REPORT does NOT use OBS numbers like PROC PRINT. If you MUST have obs numbers, then just create them in a DATA step, they will be more accurate. Before you jump straight to macro code, always start with a working program and then macro-ize it. So I would recommend making sure the OBS column is right before macro-izing. This worked for me to make sure I had a working SAS program -- I used a DATA step to capture _n_ and in a production program, I would sort sashelp.cars first:

proc sort data=sashelp.cars out=cars_make;
  by make;
run;
   
data finalcars;
   set cars_make;
      ** capture internal obs number into variable for reporting;
      ** after ensuring that data is correctly ordered by Make;
     _obs = _n_;
run;
 
** using firstobs and obs does not make sense if you are wanting to "group" by make;
** but I kept that part in the report definition;
proc report data=finalcars (firstobs= 1 obs= 50) ;
   column make _obs DISP_make _all_;
   define make /order noprint;
   define _obs /order;
   define DISP_make /computed;
   compute before make;
      cnt_make + 1;
      hold_make= make;
   endcomp;
   compute make;
      length sval $100 bk $20;
      if mod(cnt_make, 2) = 0 then bk= 'cxc4efd3'; 
      else bk= 'cxdddddd';
      sval= catt('style={background=', bk, '}');
      call define (_row_, 'style', sval);
   endcomp;
   compute DISP_make / character length= 20;
      DISP_make = hold_make;
   endcomp;
run;

 

Cynthia

View solution in original post

7 REPLIES 7
jvdl
Obsidian | Level 7

Try swapping:

compute obs;
	_obs + 1;
	obs = _obs;
endcomp;

For:

compute obs;
	obs = _obs;
	_obs + 1;
endcomp;
braam
Quartz | Level 8
@jvdl
Thanks! That works.

One more related question: In such a code, Is 1 initially assigned to _obs? Because its initial value hasn't been given.
braam
Quartz | Level 8

Just noticed a weird pattern in the column of obs:

8 and 28 are missing.

 

Could anybody help me with that?

Cynthia_sas
SAS Super FREQ

Hi:

  Keep in mind that the "fake" obs number you are computing will NOT necessarily match the obs number that PROC PRINT would show.

  Why are you overcomplicating this by using a macro. The best practice is to get the program working for 1 set of hardcoded variables and THEN to make the program more general by adding macro logic. It seems like you did NOT start with a working program.

 

Cynthia

Cynthia_sas
SAS Super FREQ

Hi:

  The reason you are getting "odd" calculated values for your fake obs numbers is because PROC REPORT is calculating your value on EVERY row it writes and EVERY break. So you have a header row, that is row #1 and technically, the first data row on the report is absolute row #2. And, PROC REPORT is doing something at the break for MAKE so that is throwing off your COMPUTE block.

 

  PROC REPORT does NOT use OBS numbers like PROC PRINT. If you MUST have obs numbers, then just create them in a DATA step, they will be more accurate. Before you jump straight to macro code, always start with a working program and then macro-ize it. So I would recommend making sure the OBS column is right before macro-izing. This worked for me to make sure I had a working SAS program -- I used a DATA step to capture _n_ and in a production program, I would sort sashelp.cars first:

proc sort data=sashelp.cars out=cars_make;
  by make;
run;
   
data finalcars;
   set cars_make;
      ** capture internal obs number into variable for reporting;
      ** after ensuring that data is correctly ordered by Make;
     _obs = _n_;
run;
 
** using firstobs and obs does not make sense if you are wanting to "group" by make;
** but I kept that part in the report definition;
proc report data=finalcars (firstobs= 1 obs= 50) ;
   column make _obs DISP_make _all_;
   define make /order noprint;
   define _obs /order;
   define DISP_make /computed;
   compute before make;
      cnt_make + 1;
      hold_make= make;
   endcomp;
   compute make;
      length sval $100 bk $20;
      if mod(cnt_make, 2) = 0 then bk= 'cxc4efd3'; 
      else bk= 'cxdddddd';
      sval= catt('style={background=', bk, '}');
      call define (_row_, 'style', sval);
   endcomp;
   compute DISP_make / character length= 20;
      DISP_make = hold_make;
   endcomp;
run;

 

Cynthia

braam
Quartz | Level 8
@cynthia

Thanks a lot for your kind explanation. I would follow your suggestion.

Regarding FIRSTOBS and OBS, I am having them because I just would like to see the first 50 observations. I hope that they do not raise up any issues. (Well, PROC REPORT is still strange to me...)
Cynthia_sas
SAS Super FREQ
Hi:
FIRSTOBS and OBS work at the data set level, returning rows BEFORE they go to PROC REPORT. Trying to duplicate the internal observation number with PROC REPORT isn't something I would recommend trying in a COMPUTE block for the reasons previously mentioned.
Cynthia

SAS Innovate 2025: Register Now

Registration is now open for SAS Innovate 2025 , our biggest and most exciting global event of the year! Join us in Orlando, FL, May 6-9.
Sign up by Dec. 31 to get the 2024 rate of just $495.
Register now!

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

SAS Training: Just a Click Away

 Ready to level-up your skills? Choose your own adventure.

Browse our catalog!

Discussion stats
  • 7 replies
  • 3294 views
  • 0 likes
  • 3 in conversation