A Data Science Central Community
If anyone can provide me with an answer to the following conundrum i will be eternally grateful.
I have written a series of programs in SAS to optimise 26 campaigns for a client’s Direct mail activity in 2011,
However this relies on the ability to run multiple scenarios to identify the ‘best’ in terms of # of new customers and roi.
If I use the Ranui function within sas, then I effectively get a normal or skewed random number e.g. the chance of every customer being assigned the same campaign exists, but the likelihood of the sas random number generator actually assigning this scenario is non-existent
Any ideas as to how I can generate ‘true’ random number within SAS ?
it sounds to me like you don't really want random selection (customer will be targeted - but 100% sure you are only in unlimited # of runs).
You may add some extra logic to control your selection process. E.g. put clients not selected by random in last 5 rounds into 26th segment, etc.