About WendyCzika

WendyCzika · ‎01-04-2016

For question 1), you can attach multiple Input Data nodes to a modeling node (or to a Control Point node then modeling node as you mention), but you can only have at most one data set of each role connecting to the modeling node. So you can have a data set with the Role property set to Raw or Train in the Input Data node, one with role Validate, and one with role Test all feeding into a modeling node, and the modeling node will use the data sets according to their role, score each one, and export the scored versions to subsequent nodes, keeping the same role. Hope that makes sense and helps.

WendyCzika · ‎01-04-2016

Please also check out the links from the Getting Started tab for SAS Enterprise Miner from the SAS support site : http://support.sas.com/software/products/miner/index.html#s1=2 and also the Documentation tab.

WendyCzika · ‎12-13-2015

The SCALE= option isn't available in PROC HPGENSELECT which is used in the HP GLM node, but you could use the Negative Binomial distribution instead of the Poisson to help with overdispersion.

WendyCzika · ‎12-09-2015

Oh great, then you can use the HP GLM node (on the HPDM tab) - set the level of your target as interval (even though it is integer-valued) and by default, the HP GLM node will use the Poisson probability distribution with a log link function.

WendyCzika · ‎12-09-2015

Hi there... 1) Yes, decision trees make no assumptions about the distribution of your target and are very robust, so you can certainly use that for variable selection 2) That depends... what version of Enterprise Miner do you have?

WendyCzika · ‎12-01-2015

Have you seen this video yet? https://youtu.be/OKTL6R5lCk4

WendyCzika · ‎12-01-2015

In the Regression node, you can set Use Selection Defaults to No, then click on the ellipsis for Selection Options. There you can change Hierarchy Effects to All then select Single for the Moving Effect Rule property. Hope that does the trick!

WendyCzika · ‎12-01-2015

I'm guessing those terms you are seeing that include 2 inputs are actually polynomial terms (those are "interactions" of the interval inputs), so you need to set Polynomial Terms to No as well if you don't want any interaction effects. The Two-Factor Interactions property applies to interactions involving the class inputs only.

WendyCzika · ‎12-01-2015

You would have to calculate this in a SAS Code node. If you connect your Regression node to a SAS Code node, then in the SAS Code node, the data set represented by the &em_import_data macro variable contains your actual and predicted target for each obervation. So you could do something like: data calc_mape; set &em_import_data; mape = %EM_RESIDUAL / %EM_Target; run; proc means data=calc_mape sum; var mape; run;

WendyCzika · ‎11-23-2015

The Link Analysis node first became available in SAS Enterprise Miner 12.3 (on SAS 9.4).

WendyCzika · ‎11-20-2015

Even if you don't have any missing values in your training data, the Decision Tree node is going to include missing values in the rules it creates in case missing values are encountered when scoring new data. And you can specify how they should be handled with the Missing Values property of the Decision Tree node. Hope that clears up any confusion!

WendyCzika · ‎10-29-2015

SAS Model Manager can handle all that you are asking for, please see http://supporttest.unx.sas.com/software/products/modelmgr/index.html for more info.

WendyCzika · ‎10-28-2015

Yes, you can do that. You also might want to check out this tip on other ways you can share metadata across projects, users, and data sources: Tip: Defining Global Metadata for SAS® Enterprise Miner™ Projects Hope that helps! Wendy

WendyCzika · ‎10-26-2015

It might be better if you override the scorecard points directly from the Scorecard node so you have more control. You can do this from the interactive editor for Scorecard Points from the property sheet for the Scorecard node: Hope that helps, Wendy

WendyCzika · ‎10-23-2015

I'm not sure why that would be happening. Do you see other plots in the Results, like the Effects Plot? Do you see the Score Rankings Overlay plot if you run with one of the sample data sources available under Help?

Online Status	Offline
Date Last Visited	a week ago