11-17-2016 06:55 AM
I hope to know about the different algorithms used internally by SAS to generate random sequences according to a given distribution (RANUNI function--> uniform distribution). If there are any geeks around, are there any papers that adresses the algorithms used by SAS ?
Thank you so much for your help.
11-17-2016 07:51 AM
The main thing you need to know about the RANUNI, RANNOR, etc, set of functions is that they were written in the 1970s and are not recommended for serious simulation or Monte Carlo computations. See the article "Six reasons you should stop using the RANUNI function to generate random numbers." The article explains differences between the older linear congruential routines and the more modern Mersenne Twister (MT) algorithm that is used by the RAND function. The documentation for the RAND function has references to the scholarly papers where you can learn more about MT and random number generation.
If you are interested in simulating statistical data in SAS, you might be interested in the paper "Ten Tips for Simulating Data with SAS," which includes tips and techniques for efficient simulaiton as well as references for the serious researcher.