Hello, I have created two models in Enterprise Miner i.e. a regression model and a neural network model. I have also input a profit matrix as follows: Solicit a good customer: A Solicit a bad customer: B Ignore a good customer: C Ignore a bad customer: D             Solicit  Ignore Good      A          C Bad        B          D The results of the model comparison node (average profit) state that the regression model is better. I guess the formula for concluding to the average profit is: (TP*A+TN*D+FN*C+FP*B)/N. So the regression must have a bigger average profit. My question is about the additional assessment criteria and more specifically the cumulative total expected profit chart. In this chart, the curve for the neural network is above the curve for the regression in all the deciles of the horizontal axis. How come then regression outperforms neural networls in the average profit measure? Should the two results be aligned? Thank you very much in advance, Andreas ... View more

Hello, I have created a user defined format in Enterprise Guide and I saved it in a permanent library. I know from Base SAS that in order to tell SAS to search in a particular library so as to find a format we use the fmtsearch option. How can we apply the fmtsearch option using thew GUI? Thank you very much in advance, Andreas ... View more

Hello, I have a hierarchical data set (three levels of hierarchy - Whole Company/ Product Type/ Product Item/ Region) ) that contains a dependent variable (demand) an independent variable (price) and a binary independent variable (BOGOF). For the variable price i have set the aggregation and accumulation options to "average" and for the BOGOF variable i have set them to "maximum value". The forecasts are produced automatically and then i select one node from the bottom level of the hierarchy to do some scenario analysis based on the binary independent variable. The system has extrapolated into the future all the variables including the binary independent variable. The result is that the extrapolated values of the BOGOF variable take decimal values between 0 and 1 which do not make any sense. The desired result would be that the extrapolated values of the BOGOF variable take the value of 0 and so as the analyst could make what if scenarios by chnaging some of them to 1. Is there a way to tell the system not to extrapolate a specific independent variable but to set its future values to 0? Thanks in advance, Andreas ... View more

Hello, I want to create a data set in SAS Enterprise Guide with data drawn from a questionnaire. I also want to type in some user defined missing values. I know that in the Table Editor of Base SAS i can insert up to 28 user defined missing values by typing ._ or .a or .b etc until .z. These values would appear in the data set without the periods but they will be treated as missing values. I want to do the same thing in EG by typing the user defined missing values but it does not seem to happen. When i type in e.g. .a in a cell of a numeric variable i get an error that this value is innapropriate. Thanks in advance, Andreas ... View more

Hello, I have the following macro: %macro varscope; data _null_; set xxx.yyy end=final; call symputx('localtype'||left(_n_), Customer_Type); if final then call symputx('localn',_n_); run; %put _user_; %mend varscope; %varscope; In the log i see that the macro variables are stored in the global symbol table. Why is that? According to the theory if a macro variable is created withing a specific macro like the varscope it should be placed in a local symbol table. Why in this case the macro variables are storeed in the global symbol table? Thanks in advance, Andreas ... View more