I'd like to evaluate conditions that are stored in a character vector with an IF clause.   Ideally, the code below would print 2 and 3, but you'll noticed I used the vague denominator SomeFunction when evaluating the condition. My question is: Is there such a function, and if so, what's it named ?   proc iml; x = {1, 2 , 3, 4}; cond = {"x[i] = 0", "x[i] > 0", "x[i] < 10", "x[i] = 10"}; do i = 1 to 4; if SomeFunction(cond[i]) then do; print i; end; end; quit;     If there is a 'cleaner' aproach to this, please let me know, I am not very familiar with IML so any advice is welcome.   ... View more

On each iteration of the data step, the macro variable should change it's value to the current value of a column in the data and then use it in further calculations. The example below shos what I would like to do - how do I set the value of 'macrovar'? This example is overly simplified, the real script I am working on doesn't really have any other solution (that I can think of).     data have; input Var1 Var2; datalines; 4 10 7 20 9 30 3 40 2 50 5 60 run; data want; set have; call symputx('macrovar',Var1); Var3 = Var2 + &macrovar. ; /*equivalent of Var3 = Var2 + Var1*/ run;   ... View more

I am trying to understand how to account for oversampling in Enterprise Miner - I did find and read some related topics, however I am still at a loss. I am trying to predict a binary event that has a probability of 4%. I've went through the following three scenarios: 1. Oversample the data in SAS BASE so that the event (TARGET=1) is 30% of the sample. Did not create any decisions in EM, ran a HP Forest and got some decent results on the test / validation data sets. 2. Oversample in SAS BASE to 30%, then setup decisions in EM: in the Prior Probabilities tab (Data Source Wizard, if that matters) I put 0.04 for Level 1 and 0.96 for Level 0 in Adjusted Priors. The HOP Forest Model produced some bizarre results - all predictions were TARGET = 0, so not a single case with TARGET=1 was predicted correctly. Moreover, the ROC index for test data was 0.88 and the Cumulative Lift for the 10th percentile was 6.2 - these values seems to indicate a good model, which is obviously not the case, 3. No oversampling done, no decisions. I got the exact same results as point 2 above - what's the point of prior probabilities then ?   What am I doing wrong? I thought scenario 2 would be the way to go, but I am making a mistake and I can't figure out what it is. ... View more