When starting your ods output, specify all necessary physical and logical paths: ods html file=outfile1 path="&outpath./" (url="http://XXX/~%sysget(USER)/&urlpath./") gpath="&outpath./&graphbase./" (url="http://XXX/~%sysget(USER)/&urlpath./&graphbase./") ; In this example (taken from one of our batchjobs), path and gpath contain the physical file/path names where the files are written, while the url= options name the logical paths that are written into the files when creating the links to sub-pages and graphics. This makes sure that all graphics show up correctly when the main page is accessed by browser. So imagine we want to create a HTML report called myreport, and the job is run by user itsme on server xxx.com: %let outpath=$HOME/public_html/myreport; %let graphbase=graphics; %let urlpath=myreport; filename outfile1 "&outpath./myreport.html"; /* ods statement with resolved values */ ods html file="/home/itsme/public_html/myreport/myreport.html" path="/home/itsme/public_html/myreport/" (url="http://xxx.com/~itsme/myreport/") gpath="/home/itsme/public_html/myreport/graphics/" (url="http://xxx.com/~itsme/myreport/graphics/") ; When you now access http://xxx.com/~itsme/myreport/myreport.html, all graphics will appear nicely, and links to sub-pages (eg from proc report) will also work. ... View more

Hello ManOfHonor-   The Interactive Grouping node property Allow Out-of-Range Values is set to No by default.  Here is information from the tooltip.   Allow Out-of-Range Values Specifies whether values outside of the user-specified minimum and maximum values for scaling are allowed. When the Weighting Method is selected for the Binary Transformation, scaled values greater than 1 are set to 1 and values less than 0 are set to 0 for creating the Frequency variable.   For details about how to proceed, from within Enterprise Miner select Help -> Contents -> SAS Credit Scoring -> Interactive Grouping Node -> Interactive Grouping Node Train Properties: Interval Target Options.  That section tells about various properties that you can set, depending on what you decide is appropriate for your analysis.   Have a great day! ... View more

When you have the project selected in the project panel, click on the "..." for Project Macro Variables, see screenshot. ... View more

Unfortunately, there is only so much stability you can get out of the small number of observations you are working with.   I also would be concerned that there would be great variability in the model metrics for Train/Validate for slight changes in the model (again, a stability issue).  Inflating the frequency improves the calculated percentage of events but it doesn't really represent more information to the model.    You have few enough observations that I would consider not splitting at all and then doing some type of bootstrap modeling to try and get at a small set of predictors for fitting a model.  It also might be useful to consider a one-split Tree model to see if there are any of the nodes that almost never have the event.  You could then attempt to do your analysis on the remaining nodes which would by definition have a higher percentage of the event of interest.  In this way, you could pre-filter the data in a second flow where you could model the observations having a higher chance of the target event.  This could provide a better solution than fitting against the whole data set given how rare your event is.   You would have to deploy the score code conditionally depending on whether a given observation was pre-filtered or whether it was subject to be scored by the Tree model.     Again, this doesn't solve your problem with having so few events to begin with but you might find it easier to model the remaining observations after removing those with almost no chance of having the target event.   SAS Enterprise Miner was designed for data mining data sets which are often extremely large, so this would require a bit more work but should be easy to implement.   Hope this helps! Doug  ... View more

Sorry . I never use EM. but the following could oversample.   proc surveyselect data=have sampsize=(1000 1000) out=data_oversample; strata good_bad; run;   Note: assuming you have 1000 bad obs. ... View more