Hi,
Ever since I migrated to a new SAS server, i notice that random number generation for a seed value [334766080] gets different in this new server.
Code
data A(drop=i);
call streaminit(334766080);
do i = 1 to 5;
x123 = rand("uniform"); output;
end;
run;
Results In OLD server
x123
0.7852188905
0.1264403092
0.5529167121
0.8073481501
0.3563103683
Results in NEW Server
x123
0.9406947785
0.9712292252
0.0413521428
0.5594002416
0.1823783112
Seed value is exactly the same. Is there something else which affects the random number generation process?
Thanks in advance!
Dilip
It seems that there is a change in random number generation from SAS/STAT14.1
When I added following option, I get old results 🙂
OPTIONS SET=SAS_RNG_METHOD=MT1998;
Thanks,
Dilip
What is the difference between your "OLD" and "NEW" server?
A test with SAS 9.4 TS1M2 on AIX gave me your "OLD" results; a test with 9.2 on the same platform also.
Hi Kurt,
Sorry- i forgot to mention that.
OLD server= SAS9.3
New Server= 9.4 (TS1M3 MBCS3170)
Thanks,
Dilip
It seems that there is a change in random number generation from SAS/STAT14.1
When I added following option, I get old results 🙂
OPTIONS SET=SAS_RNG_METHOD=MT1998;
Thanks,
Dilip
Since it does seem to affect not only certain SAS/STAT procedures, but the random function in Base SAS, this is good to know.
It also explains why I got the "old" numbers. Our SAS/STAT is at 13.2.
Don't miss out on SAS Innovate - Register now for the FREE Livestream!
Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.
ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.
Find more tutorials on the SAS Users YouTube channel.