Great - thanks - will check this out but not sure if I have necessary access.. ... View more

Thanks for your reply! Apologies for not replying sooner - I need to set up the alert features on the forum! We have ended up creating a limited selection of datasets... ... View more

PROC PRODUCT STATUS says 9.21_M2 for SAS/GRAPH and 9.21_M3 for 'Base Product' if that makes sense? ... View more

Fred, You can test your model on a small random sample of the customers who only use online banking solution to see if the model is applicable to that cohort. As a practical matter, models that are deployed for the 'right' segments can go wrong for all sorts of reason. A small scale, pre-test like this should take out potential complications embedded in full scale, 'live' deployment. Models are supposed to be deployed for the intended segment. This is essential for model performance measurement. This does not mean the model has little or no applicability onto segments that do not overlap much with the original model universe. If a model is driven by variables that are shared between online banking and non-online banking customer bases, the model may very well stand up OK. What you can test is in building the model, create a flag or a set of variables to 'single out' those on banking customers only to see if such flag is predictively significant or will bias your model in any way. In some cases, such flag is significantly statistically but does not call for separate model for each. Sometime the flag may behave like a 'fault' that separates continents, when a separate model may be good idea. At variable selection phase, you may pay attention to variables that are unique to one segment but not to the rest. Best Regards Jason Xin ... View more

I will certainly test for customer length as a predictor. I guess my concern was that with two customers A and B having different amounts of historic data it would be wrong for a particular predictor not to consider this. The business problem I am trying to solve is to identify existing customers with a higher likelyhodd of purchasing a particular product. As such I can probably defend excluding customers with limited history and what I have done is to 'require' that all customers in my observation data has been with us for a required number of months. This, of course, will lead to some bias but hopefully the model will still have some predictive power THanks for your help Chris ... View more

Sorry Gregor I was a bit quick off the mark there. Cannot get this to work for neither DATA or PROC SQL the way I thought it would.. ... View more

The basic idea with Proc Tabulate is you can specify the display format with the selected statistic. Part of an example: class classvar; var var1; table classvar,var1*(mean*f=comma6.1 std*f=f6.4 min*f=mycustomformat.); ... View more

All PROCs that I've came across that generates data output, the result is a table. If it's essential for you to have a view as output, you can probably mimic your PROC TRANSPOSE using a data step view. /Linus ... View more

And here's the answer from the helpdesk, using the "Log4net" Apache dll, you can modify the File parameter in the logging.config file; ... View more

There are some tricks in 3.1 that you can do. Maybe email the pdf so the user can quickly view the results that you are referring to? The SAS Paper: http://www2.sas.com/proceedings/forum2008/064-2008.pdf Provides some interesting steps to distributing reports from Web Report Studio to elsewhere in your environment. (But at least you have 1 more reason to push towards a 9.2 / 4.2 implementation 🙂 ) Angela Hall Zencos Consulting http://www.zencos.com ... View more

I am pleased! ... View more

Cheers Cynthia - will def give that a go 🙂 ... View more