you need to change "TARGET_VARIABLE" for your target variable... is the name of your target "TARGET"? Then your model statement should look like this: MODEL TARGET= %EM_INTERVAL_INPUT / VIF TOL COLLINOINT; Alternatively you can use one of these macros that resolves to your target. %EM_TARGET %EM_BINARY_TARGET %EM_ORDINAL_TARGET %EM_NOMINAL_TARGET Give it a try and if it does not work for you, please include the log so that we can google or check in SAS documentation what you are missing. ... View more

Hi Omer, I think what you need is to calculate VIF to assess multicollinearity. HPReg node this automatically for you. If you are working with an older EM version you can also write some code for proc reg. See https://communities.sas.com/message/188233 Since you are talking about trees, they are useful to detect 2-way interactions. After your first split, what variable is selected for the largest tree of maximum depth 2? Do you get similar results from both VIF and a tree to detect 2-way interactions? I hope that helps, Miguel ... View more

, Here is a repository with real data sets (UCI Machine Learning Repository: Data Sets). Plot your target with some of their inputs and then try to capture that relation with a variable transformation. Try the transformations in these examples, as well as the ones described in the help for the Transform Variables node. The section "Best Power Transformations" is really useful... Not 100% sure, but I think this is a real data set. Good luck! -Miguel ... View more

Hi Gabee, You might be missing to add your variables to the data set &em_user_emscorevar. Take a look at Susan's examples in this article . Did that fix it? Good luck! -Miguel ... View more

Kyvas, also check out these videos! these are a really cool resource too! Knowledge Base Video Portal http://support.sas.com/rnd/app/video/index.html Thanks, Miguel ... View more

Hi Joel, SVM node was experimental and the proc that runs behind the scenes does not produce score code. This is what makes the Ensemble node fail. Starting EM13.2 you have HPSVM node that both produces great results and also plays well with the Ensemble node. This is the very first time I added an HPSVM in my ensembles. It has good discordance and is helping my gradient boosting. I will keep track of how well HPSVMs complement other types of models. Sorry SVM is not a good candidate for your ensembles. I hope you get to work with HPSVM soon! Good luck, Miguel ... View more

Kyvas, SAS is the #1tool to do Analytics. And it is also a great time to learn it no matter your background... SAS University is offering the SAS programming essentials for free. I have seen people from all walks of life master SAS basics in matter of weeks. My advice: get familiarized with SAS through SAS University and then decide which learning path is right for you. To give you a quick example, two of my favorite SAS products are SAS Enterprise Miner and SAS Forecast Studio. Enterprise Miner is the best tool to do data mining and data science. It took me from zero to hero in Analytics in no time. Forecast server is my go-to for forecasting, and it makes it really easy for me to analyze time series data even if I am not a statistician and I know almost nothing about econometrics. At this point it is really hard to know if Enterprise Miner or Forecast Server would be the next step for you after you are familiar with SAS (aka Base SAS or SAS Studio). You can decide your next step later. For now check out SAS University, and SAS white papers and demos all over the internet. Try googling SAS and any industry, any problem, and I am pretty sure you will find something really good. In the meantime, you might find useful: SAS Tutorials | SAS Training SAS Training - Find a Course: Learning Paths A strong advantage for you as an MBA is that you are familiar with solving problems in a case scenario. Most of my MBA friends know what analysis they need in order to make a decision. That's great! Folks like me build more analysis than we need to... Sometimes just to make sure... More information on SAS U download for independent learners in this link: SAS Analytics U | SAS Connect with me on linkedin and I will suggest you some SAS products, papers/tutorials and solutions that might be particularly interesting. Good luck, Miguel ... View more