I'm trying to run this code in SAS
proc mi data=Mono1 seed=1423741 nimpute=6 out=outex18;
class Trt;
monotone reg;
mnar adjust( y1 / adjustobs=(Trt='0') parms(shift=shift_c)=parm1);
adjust( y1 / adjustobs=(Trt='1') parms(shift=shift_t)=parm1);
var Trt y0 y1;
run;
but I get the following error
ERROR: The _Imputation value is not a valid SAS name.
NOTE: The SAS System stopped processing this step because of errors.
WARNING: The data set WORK.OUTMI was only partially opened and will not be saved.
also mnar is in red.
How can I solve this problem?
the data is
data Mono1;
input trt y0 y1;
cards;
0 10.5212 11.3604
0 8.5871 8.5178
0 9.3274 .
0 9.7519 .
0 9.3495 9.4369
1 11.5192 13.2344
1 10.7841 .
1 9.7717 10.9407
1 10.1455 10.8279
1 8.2463 9.6844
;
run;
Good day,
It appears that you example is from documentation. usually it helps when taking the whole bit from the example. There seems to be a typo in parameters file. This works with SAS 9.4:
data Mono1;
input Obs Trt y0 y1;
cards;
1 0 10.5212 11.3604
2 0 8.5871 8.5178
3 0 9.3274 .
4 0 9.7519 .
5 0 9.3495 9.4369
6 1 11.5192 13.1344
7 1 10.7841 .
8 1 9.7717 10.8407
9 1 10.1455 10.7279
10 1 8.2463 9.5844
;
run;
data parm1;
input Obs _IMPUTATION_ SHIFT_C shift_t;
cards;
1 1 -0.56986 -0.90494
2 2 -0.38681 -0.84523
3 3 -0.58342 -0.92793
4 4 -0.48210 -0.99031
5 5 -0.57188 -1.02095
6 6 -0.57604 -1.008
;run;
proc mi data=Mono1 seed=1423741 nimpute=6 out=outex18;
class Trt;
monotone reg;
mnar adjust( y1 / adjustobs=(Trt='0') parms(shift=SHIFT_C)=parm1)
adjust( y1 / adjustobs=(Trt='1') parms(shift=shift_t)=parm1);
var Trt y0 y1;
run;
Related
Suppose i have random diagnostic codes, such as 001, v58, ..., 142,.. How can I construct columns from the codes which is 1 for the records?
Input:
id found code
1 1 001
2 0 v58
3 1 v58
4 1 003
5 0 v58
......
......
15000 0 v58
Output:
id code_001 code_v58 code_003 .......
1 1 0 0
2 0 0 0
3 0 1 0
4 1 0 0
5 0 0 0
.........
.........
You will want to TRANSPOSE the values and name the pivoted columns according to data (value of code) with an ID statement.
Example:
In real world data it is often the case that missing diagnoses will be flagged zero, and that has to be done in a subsequent step.
data have;
input id found code $;
datalines;
1 1 001
2 0 v58
2 1 003 /* second diagnosis result for patient 2 */
3 1 v58
4 1 003
5 0 v58
;
proc transpose data=have out=want(drop=_name_) prefix=code_;
by id;
id code; * column name becomes <prefix><code>;
var found;
run;
* missing occurs when an id was not diagnosed with a code;
* if that means the flag should be zero (for logistic modeling perhaps)
* the missings need to be changed to zeroes;
data want;
set want;
array codes code_:;
do _n_ = 1 to dim(codes); /* repurpose automatic variable _n_ for loop index */
if missing(codes(_n_)) then codes(_n_) = 0;
end;
run;
I have data set,
CustID Rating
1 A
1 A
1 B
2 A
2 B
2 C
2 D
3 X
3 X
3 Z
4 Y
4 Y
5 M
6 N
7 O
8 U
8 T
8 U
And expecting Output
CustID Rating ID
1 A 1
1 A 1
1 B 1
2 A 1
2 B 2
2 C 3
2 D 4
3 X 1
3 X 1
3 Z 2
4 Y 1
4 Y 1
5 M 1
6 N 1
7 O 1
8 U 1
8 T 2
8 U 1
In the solution below, I selected the distinct possible ratings into a macro variable to be used in an array statement. These distinct values are then searched in the ratings tolumn to return the number assigned at each successful find.
You can avoid the macro statement in this case by replacing the %sysfunc by 3 (the number of distinct ratings, if you know it before hand). But the %sysfunc statement helps resolve this in case you don't know.
data have;
input CustomerID Rating $;
cards;
1 A
1 A
1 B
2 A
2 A
3 A
3 A
3 B
3 C
;
run;
proc sql noprint;
select distinct quote(strip(rating)) into :list separated by ' '
from have
order by 1;
%put &list.;
quit;
If you know the number before hand:
data want;
set have;
array num(3) $ _temporary_ (&list.);
do i = 1 to dim(num);
if findw(rating,num(i),'tips')>0 then id = i;
end;
drop i;
run;
Otherwise:
%macro Y;
data want;
set have;
array num(%sysfunc(countw(&list., %str( )))) $ _temporary_ (&list.);
do i = 1 to dim(num);
if findw(rating,num(i),'tips')>0 then id = i;
end;
drop i;
run;
%mend;
%Y;
The output:
Obs CustomerID Rating id
1 1 A 1
2 1 A 1
3 1 B 2
4 2 A 1
5 2 A 1
6 3 A 1
7 3 A 1
8 3 B 2
9 3 C 3
Assuming data is sorted by customerid and rating (as in the original unedited question). Is the following what you want:
data want;
set have;
by customerid rating;
if first.customerid then
id = 0;
if first.rating then
id + 1;
run;
I would like to create a table that has three variables where var2 is a percentage of var1 and var3 is a percentage of var 2, broken down by class variables that have missing values.
To explain, imagine I have data showing who applied, was interviewed, and was hired for a job, e.g.
data job;
input applied interviewed hired;
datalines;
1 1 1
1 1 1
1 1 1
1 1 0
1 1 0
1 1 0
1 0 .
1 0 .
1 0 .
1 0 .
;
run;
it's very easy to create a table that shows the count of who applied, and then the percentage of those who were interviewed and then of those people, the percentage who was hired.
proc tabulate data = job;
var applied interviewed hired;
tables applied * n (interviewed hired) * mean * f=percent6.;
run;
which gives:
applied interviewed hired
10 60% 50%
Now I would like to break that down by several class variables with missing values.
data have;
input sex degree exp applied interviewed hired;
datalines;
0 1 1 1 1 1
1 . 0 1 1 1
. 0 1 1 1 1
0 1 0 1 1 0
1 0 1 1 1 0
0 1 0 1 1 0
1 . 1 1 0 .
0 1 . 1 0 .
. 0 0 1 0 .
1 0 0 1 0 .
;
run;
If I do one class variable at a time it will give me the correct percentages:
proc tabulate data = have format = 6.;
class sex;
var applied interviewed hired;
tables sex, applied * sum (interviewed hired) * mean * f=percent6.;
run;
Is there a way to do all three class variables in the table at once and get the right percentage for each category. so the table looks like:
applied interviewed hired
sex
0 4 75% 33%
1 4 50% 50%
degree
0 4 50% 50%
1 4 75% 33%
exp
0 5 60% 33%
1 4 75% 67%
This is something I must do many, many times and I need to populate tables in a report with the numbers, so I'm looking for a solution where the table can be printed all in one step.
How would you solve this problem?
The problem you're running into is that of missing data. When a case is missing for any class variable, it is eliminated from the entire table, unless you specify MISSING in the proc call. So, for example, your 4th sex=0 who did not interview was missing EXP; so they didn't show up at all in the table, though you would want them showing up in SEX.
You can get the correct numbers, mostly:
proc tabulate data = have format = 6. missing;
class sex degree exp;
var applied interviewed hired;
tables (sex degree exp), applied * sum (interviewed hired) * mean * f=percent6.;
run;
However, you have an extra row that includes those with missing data. You cannot eliminate those rows from the printed output while also including them in the other class calculations; this is just one of those limitations of SAS tabulation. Other PROCs have a similar problem; PROC FREQ is the only one that doesn't do this if you have multiple tables generated, but even then within one table (combined with asterisks) you will have the same issue.
The only way I've found around this is to output the table to a dataset and then filter out those rows, and PROC REPORT or PRINT or TABULATE the data back out.
I think this is close to what you want. You will have to fix the row labels, but it is one PROC TABULATE step.
title;
data have;
input sex degree exp applied interviewed hired;
datalines;
0 1 1 1 1 1
1 . 0 1 1 1
. 0 1 1 1 1
0 1 0 1 1 0
1 0 1 1 1 0
0 1 0 1 1 0
1 . 1 1 0 .
0 1 . 1 0 .
. 0 0 1 0 .
1 0 0 1 0 .
;
run;
proc print;
run;
proc summary data=have missing ;
class sex degree exp;
ways 1;
output out=stats sum(applied)= mean(interviewed hired)= / levels;
run;
data stats2;
set stats;
if n(of sex degree exp) eq 0 then delete;
run;
proc print;
run;
proc tabulate data=stats2;
class _type_ / descend;
class _level_;
var applied interviewed hired;
tables (_type_*_level_),applied*sum='N'*f=8. (interviewed hired)*sum='Percent'*f=percent6.;
run;
/**/
/* applied interviewed hired*/
/*sex */
/* 0 4 75% 33%*/
/* 1 4 50% 50%*/
/*degree */
/* 0 4 50% 50%*/
/* 1 4 75% 33%*/
/*exp */
/* 0 5 60% 33%*/
/* 1 4 75% 67%*/
I have a sample SAS code below. And I want to know how to proc report the percentage using two decimal formats. For example, I want to put 100% as just zero decimal, and all the other percentages which is not 100% with 1 decimal such like 25.0%.
Here is my code.
data a;
infile datalines missover;
input subjid trt itt safety pp complete enroll disreason;
uncomplete=(complete^=1);
datalines;
1 1 1 1 1 1 1
2 2 0 1 1 1 1
3 1 1 1 1 0 1 4
4 2 1 1 1 1 1
5 1 1 1 0 0 1 5
6 2 1 1 1 1 1
7 2 1 1 1 0 1 1
8 1 0 1 0 1 1
9 1 1 1 1 0 1 5
10 2 0 1 0 0 1 5
11 2 1 1 1 0 1 1
12 2 1 1 0 0 1 2
13 1 1 1 0 0 1 3
14 2 1 1 0 0 1 4
;
run;
data b;
set a(in=a) a(in=b);
if b then trt=3;
run;
data c;
length cat $20;
set b;
array a(6) itt safety pp complete uncomplete enroll;
array b(5) r1 r2 r3 r4 r5;
do i=1 to 6;
cat=upcase(vname(a(i)));
value=a(i);
output;
end;
do j=1 to 5;
if disreason=j then value=1;
else value=0;
cat=upcase(vname(b(j)));
output;
end;
keep trt cat value;
run;
proc format ;
value $newcat(notsorted)
'ENROLL'='Enrolled Population'
'1'='1'
'PP'='Per-Protocol Population'
'2'='2'
'ITT'='ITT Population'
'3'='3'
'SAFETY'='Safety Population'
'4'='4'
'COMPLETE'='Patients Completed'
'UNCOMPLETE'='Patients Discontinued'
'5'='Primary Reason for Discontinuation of Study Dose'
'R1'='\li360 Lack of Effect'
'R2'='\li360 Protocol Violation'
'R3'='\li360 Lost to Follow-up'
'R4'='\li360 Adverse Event'
'R5'='\li360 Personal Reason';
value trt 1='Treatment 1'
2='Treatment 2'
3='Overall';
picture pct(round)
0<-100='0009)'(prefix='(' mult=100)
0=0;
run;
option missing='' nodate nonumber orientation=landscape;
ods rtf file='c:\dispoistion.rtf';
proc report data=c completerows nowd
style(report)={frame=hsides rules=groups}
style(header)={background=white};
column cat cat2 trt, value, (sum mean);
define cat/group format=$newcat. preloadfmt order=data noprint;
define cat2/computed ' ' style(column)={cellwidth=30% PROTECTSPECIALCHARS=OFF
};
define trt/across format=trt. '' order=internal;
define value/analysis '';
define sum/ 'N' style(column)={cellwidth=35pt} style(header)={just=right};
define mean/ '(%)' format=pct. style(column)={cellwidth=35pt just=left}
style(header)={just=left};
compute cat2/char length=50;
cat2=put(cat, $newcat.);
if cat2 in ('1', '2', '3','4') then cat2='';
endcomp;
run;
ods rtf close;
You need to use a picture format:
data test;
input x;
datalines;
0.5
0.751
0.999
1.00
;;;;
run;
proc format;
picture pct100f
-1 = [PERCENT7.0]
-1<-<1 = [PERCENT7.1]
1 = [PERCENT7.0]
other=[BEST12.];
quit;
proc print data=test;
format x pct100f.;
var x;
run;
Adjust that as needed. The -1 <-< 1 means anything that is between -1 and 1, exclusive. The [PERCENT7.0] is telling it to use that format for that section.
Try using this format:
define mean/ '(%)' format=percent7.1
How can I manage proc tabulate to show the value of a variable with missing value instead of its statistic? Thanks!
For example, I want to show the value of sym. It takes value 'x' or missing value. How can I do it?
Sample code:
data test;
input tx mod bm $ yr sym $;
datalines;
1 1 a 0 x
1 2 a 0 x
1 3 a 0 x
2 1 a 0 x
2 2 a 0 x
2 3 a 0 x
3 1 a 0
3 2 a 0
3 3 a 0 x
1 1 b 0 x
1 2 b 0
1 3 b 0
1 4 b 0
1 5 b 0
2 1 b 0
2 2 b 0
2 3 b 0
2 4 b 0
2 5 b 0
3 1 b 0 x
3 2 b 0
3 3 b 0
1 1 c 0
1 2 c 0 x
1 3 c 0
2 1 c 0
2 2 c 0
2 3 c 0
3 1 c 0
3 2 c 0
3 3 c 0
1 3 a 1 x
2 3 a 1
3 3 a 1
1 3 b 1
2 3 b 1
3 3 b 1
1 3 c 1 x
2 3 c 1
3 3 c 1
;
run;
proc tabulate data=test;
class yr bm tx mod ;
var sym;
table yr*bm, tx*mod;
run;
proc tabulate data=test;
class tx mod bm yr sym;
table yr*bm, tx*mod*sym*n;
run;
That gives you ones for each SYM=x (since n=missing). That hides the rows for SYM=missing, hence you miss some values overall from your example table. (You could format the column with a format that defines 1 = 'x' easily).
proc tabulate data=test;
class tx mod bm yr;
class sym /missing;
table yr*bm, tx*mod*sym=' '*n;
run;
That gives you all of your combinations of the 4 main variables, but includes missing syms as their own column.
If you want to have your cake and eat it too, then you need to redefine SYM to be a numeric variable, so you can use it as a VAR.
proc format;
invalue ISYM
x=1
;
value FSYM
1='x';
quit;
data test;
infile datalines truncover;
input tx mod bm $ yr sym :ISYM.;
format sym FSYM.;
datalines;
1 1 a 0 x
1 2 a 0 x
1 3 a 0 x
... more lines ...
;
run;
proc tabulate data=test;
class tx mod bm yr;
var sym;
table yr*bm, tx*mod*sym*sum*f=FSYM.;
run;
All of these assume these are unique combination rows. If you start having multiples of yr*bm*tx*mod, you would have a problem here as this wouldn't give you the expected result (sum 1+1+1=3 would not give you an 'x').