Hi all,
I am trying to generate a table with 100 observations, one variable. This is a discrete variable with 4 levels so I used this code below and everything is fine.
data BMI (drop = i) ;
array prob [4] (0.01, 0.1, 0.26, 0.63);
call streaminit(1234);
do i = 1 to 100;
BMI = rand("Table", of prob
output;
end;
run;
quit;
However, what I really want is this to store the proportions (0.01, 0.1, 0.26, 0.63) in a matrix variable z, like this below, and then pass that matrix variable via array, prob
proc iml;
use CHIS.RBMI_Freq;
read all var {PERCENT} into XBMI ;
Z = ((XBMI[1:4,])*.01);
print Z;
run;
data BMI (drop = i) ;
array prob
call streaminit(1234);
do i = 1 to 100;
BMI = rand("Table", of prob
output;
end;
run;
quit;
When I do this I get an error Argument 2 to function RAND('Table',.) at line 822 column 15 is invalid.
I need help. Thanks in advance.
I don't work much with IML but I would think you could do that as a row operation in IML. Data steo is, as error says, not going to accept that syntax the way you want as Z as called will create a single variable Z which is missing.I don't know if you could instead of Print Z create a macro variable in a format that looks like the values in the first example.
BTW if you don't want to keep adding Drop i statements use Do _i_= for your loops. The paired underscores tell SAS to treat as a temporary variable and it is automatically dropped from the output.
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
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.
Ready to level-up your skills? Choose your own adventure.