Logworth Drop to 0 when Editing Rule for Interactive Decision Tree in SAS EMiner

[stechafle]

Hello. I am using the Interactive Decision Tree tool in SAS Enterprise Miner 14.3. When I Split a Node and make any edits to the rule, the logworth of that variable drop to 0 such that I cannot tell any longer its relative strength vs the other variables. Is there any workaround for this behavior?

For example, here I have a variable ALL7360 that is the 2nd most predictive that has 5 branches and a logworth near 100. 

If I remove Branch 5 and Apply, the logworth drops to 0, but I don’t think that is the correct value as there is still good variation in the Target % across the branches. The only way I have found to restore the logworth is to Prune the Node and start over. That loses any Edits made to the Rule though.

[DougWielenga]

Hello. I am using the Interactive Decision Tree tool in SAS Enterprise Miner 14.3. When I Split a Node and make any edits to the rule, the logworth of that variable drop to 0 such that I cannot tell any longer its relative strength vs the other variables. Is there any workaround for this behavior?

When you are in interactive splitting, the logworth being displayed is the logworth for that split.  If you split on a variable, it is possible that further splitting on that variable will have a logworth of zero even if the response values differ since a logworth of zero just means that the variable in question cannot be used to create a better fitting model by splitting that particular node.   To see overall variable importance (not just for that particular node), close the interactive portion of the tree and open the tree results.  Inside the tree results browser, click on Model --> Variable importance to display the overall importance values for the fitted model.  

Hope this helps!

Doug

[stechafle]

I brought up this issue with SAS Support. The resolution is that the development team confirmed the behavior and there is no known workaround at this time. They will investigate a fix in a future release of Enterprise Miner.

If you want to recreate the issue, you can follow these steps:

 - Create a new diagram.

 - Select Help -> Generate Sample Data Sources.  

   Select Home Equity (SAMPSIO.HMEQ).

 - Drag the Home Equity data source to the diagram.

 - Connect a Decision Tree node.

Go to Node 1 (the root node at the top).

   Right-click the node, choose Split Node.

 - In the Split Node window, select DELINQ, Edit Rule.

 - In the Interval Split Rule window, select Branch 5, Remove Branch, Apply, OK.