cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
M_Maldonado
‎02-28-2018
  • 338 Posts
  • 35 Likes Given
  • 23 Solutions
  • 198 Likes Received
About
Analytics Credit Scoring, Reject Inference High-Performance Data Mining Data Visualization Ensemble Models

Re: Optimal Binning in the Enterprise Miner Transform Variables node n...

by Barite | Level 11 in SAS Data Science
‎03-31-2015 11:38 AM
‎03-31-2015 11:38 AM
TMKAIG, You just made me miss my old job at credit scoring--just a little bit. I had that feeling of "I hate to love when computers beat me!". I ran the default IGN, which came with a different grouping. I was expecting it to be a little worse than our eyeball, because c'mon proc eyeball rocks... Then I manually added the alternative groupings of eyeball and smbinning, because it only takes a minute to do it on the interactive menu. Comparison results: IGN wins, eyeball is a close second, smbinning last. Here a table with the results, and screenshots as an appendix. I hope another set of eyes make sure everything is OK... I have embarrassed myself more than once on the SAS communities :smileysilly:. Metric Interactive Grouping Node (default) Eyeball SMBinning Information Value 14.201 12.279 11.991 Gini 97.756 95.513 94.872 TMKAIG, last favor. You mentioned Chi-square. I never really used it, I always used IV. Do you recommend it, yes, no, why? Also if you have data sets that I can play with, please send them my way. I frequently beat the default IGN, but sometimes it is really useful, so I never make up my mind if I should run my preferred settings vs default first. I will do that on some free time. Thanks for the interesting exercise, and I hope this helps! -Miguel Appendix 1: SAS Credit Scoring for EM - Interactive Grouping Node results 0-2000 2001-4000 4001-6500 6501-8000 8001-10000 Appendix 2: Screenshots IGN default run IGN vs eyeball IGN vs smbinning ... View more

Re: Enterprise Miner Interactive Grouping Node resets itself...

by Barite | Level 11 in SAS Data Science
‎03-31-2015 07:25 AM
‎03-31-2015 07:25 AM
Hi AWright, It sounds like you are doing things right... Give it one more try and let me know if I missed any step. 1. You drag an IG node and run it. Then you open the Interactive Grouping. 2. To manually reject variables, you select Rejected from the Variable Role Box. Then you click Apply. 3. You close and say Yes when prompted to save your groupings. 4. You change the property Use Frozen Groups to Yes. That's the way to do it... Once you are done doing your groupings and assigning variable roles, open the results. For the Output Variables, are all the "New Roles" what you would expect? Thanks, Miguel ... View more

Re: Optimal Binning in the Enterprise Miner Transform Variables node n...

by Barite | Level 11 in SAS Data Science
‎03-31-2015 07:03 AM
‎03-31-2015 07:03 AM
Hi TMKAIG1, Do you mind if you borrow the SAS code you used to generate that data set? I am really interested in optimal binning and I use a lot the Interactive Grouping Node from Credit Scoring in Enterprise Miner. Really curious to see the binning solution from the default IG node. Please share that data set (or code). Really interested in seeing some comparisons vs smbinning... Thanks! -Miguel ... View more

Re: New SAS Miner user needs help

by Barite | Level 11 in SAS Data Science
‎03-25-2015 07:13 PM
‎03-25-2015 07:13 PM
Hi MKP, Quick input to your questions, 1. I can't remember if you can switch labels for variable name. Let me get back to you on that one. 2. There are a couple ways to show you more information about the rule for that node. I really like to right click on the node and select Tools->Display English rule. 3. You can assign a variable the role of frequency. Do that when you are creating your data set, or used the Metadata node. I hope it helps, Miguel ... View more

Re: HELP in SAS E-miner

by Barite | Level 11 in SAS Data Science
‎03-25-2015 02:37 PM
1 Like
‎03-25-2015 02:37 PM
1 Like
Hi ALSUBAIE  found some good resources and posted them in this thread https://communities.sas.com/message/261213#261213. I hope that helps, Miguel ... View more

Re: Duplicate corresponding variable within a group

by Barite | Level 11 in SAS Data Science
‎03-25-2015 09:53 AM
‎03-25-2015 09:53 AM
Hi, I don't remember if we have a quick way to do this in Enterprise Miner. The quick alternative is to use proc sort. This would also work on a SAS Code Node in Enterprise Miner. Personal favorite, is to use NODUP option, and by _all_. Example: In a SAS Code in EM you might want to use proc sort data=mydata NODUP; by _all_; run; In a SAS Code Node in EM you might want to use: proc sort data=&em_import_data out=&em_export_train NODUP; by _all_; run; I hope this helps! -Miguel ... View more

Re: Getting Started with SAS Enterprise Miner

by Barite | Level 11 in SAS Data Science
‎03-24-2015 05:03 PM
1 Like
‎03-24-2015 05:03 PM
1 Like
Right, we need to make more of those! Thanks for the feedback! -Miguel ... View more

Re: How can I use the Tobit's Model in SAS?

by Barite | Level 11 in SAS Data Science
‎03-24-2015 05:02 PM
1 Like
‎03-24-2015 05:02 PM
1 Like
Hi JPI, If you truly want a censored regression (Tobit model), take a look at SAS/ETS(R) 13.2 User's Guide: The QLIM Procedure. It includes an example with the different Tobit models that you can use: Types of Tobit Models. A similar approach in SAS Enterprise Miner would be to use the Replacement Node to censor values from the right or the left of your input variables. Once you have censored your values, you can use the Regression node as usual. It would be interesting to compare the accuracy of proc QLIM and your own flow of Replacement->Regression node, but I couldn't find a paper of someone doing that. If you happen to work on that, please comment on your findings! An advantage of the second approach is that you could use HPRegression node when you are dealing with big data. I hope this helps, Miguel ... View more

Re: How can I add a lag as a regressor?

by Barite | Level 11 in SAS Data Science
‎03-24-2015 09:41 AM
3 Likes
‎03-24-2015 09:41 AM
3 Likes
Your question is how to add a lag as an input variable? I don't think any of the Modify nodes have a simple way to do it. But it is really easy to do in SAS Code. In Enterprise Guide add the below when preparing your data, or use a SAS Code Node in Enterprise Miner. EG (add this statement in any data step, and change var1 for the name of your input variable)   lag_var1= lag(var1); EM, in a SAS Code Node (change var1 for the name of your variable) data &em_export_train;   set &em_import_data;   lag_var1= lag(var1); run; I hope it helps, Miguel ... View more

Re: EM: Setting cutoff for predictive model

by Barite | Level 11 in SAS Data Science
‎03-24-2015 09:32 AM
‎03-24-2015 09:32 AM
A popular automatic cutoff selection method is to use the Event Precision Equal Recall. This method selects a cutoff where the event precision rate and the true positive rate intersect. I posted a brief example here: https://communities.sas.com/docs/DOC-6050 I hope it helps, Miguel ... View more

Re: stock price Run-up

by Barite | Level 11 in SAS Data Science
‎03-20-2015 12:17 AM
‎03-20-2015 12:17 AM
Lenon, Below a data mining approach that you might want to try. If you want to go into time series analysis, you probably more about it than I do! It's been a while... Data mining approach Back in the day in grad school at the Institute for Advanced Analytics one of our teachers described a really neat idea: Use a linear regression procedure to output the parameter estimates (betas) of a trend for a 1-day, 2-day, 3-day, and nth-day prediction (or hour for that matter). Then use those betas as inputs for a predictive model that predicts the nth+1 beta. If the nth+1 beta is positive then you have a positive run-up that is quantifiable in terms of both the magnitude and also the accuracy coming out of your best predictive model.          [If someone knows the formal name of this technique, please comment about it! really interested!!!] Lenon, give it a try with this approach. If you run into trouble, happy to dig up that code or try to reproduce it. Good luck! Miguel ... View more

Re: How create a ROC-curve for K-means clustering (K>2) with SAS ENTRI...

by Barite | Level 11 in SAS Data Science
‎03-19-2015 08:44 AM
‎03-19-2015 08:44 AM
Hey Noel, ROC curves help you to assess models for a binary target (sensitivity and 1-specificity connect the dots of true positives and false negatives). Clustering is unsupervised learning, which means you don't need a target. This is why you don't see ROC curves if you don't have a target that can be assessed in terms of sensitivity and 1-specificity. I hope it helps, M ... View more

Re: How to combine nodes in Enterprise miner into one model node?

by Barite | Level 11 in SAS Data Science
‎03-15-2015 01:11 AM
‎03-15-2015 01:11 AM
It sounds like you want to create your own "template" of a diagram flow. Right click on the name of your diagram and select "Save as" to save it as an XML. To import this diagram and use it on new data, right click on "Diagrams", then select import from XML and the template you created. I hope it helps, Miguel ... View more

Re: Cross-Post From Text Mining Community Forum - Newbie Question

by Barite | Level 11 in SAS Data Science
‎03-12-2015 09:50 AM
3 Likes
‎03-12-2015 09:50 AM
3 Likes
Hey Zachary, Did you get what you needed from the Text Mining community? There are a couple ways to do this. Since you have your file already delimited by pipes, you can use proc import with a dlm='|' on the infile option. E.g. (change C:\Users\test.txt for the path and name of your data ) infile 'C:\Users\test.txt' dsd dlm='|'; I grabbed this example from Indiana University's Knowledge base: Data test;   Infile "C:\Users\test.txt" DLM='|' DSD LRECL=400 FIRSTOBS=2;   Input ID Age Sex $ Income Race $ Hight Weight;   run; Source: https://kb.iu.edu/d/bcjf More info about reading delimited text files here: http://support.sas.com/techsup/technote/ts673.pdf I hope this helps, Miguel ... View more

Re: Deploy Model Results to web

by Barite | Level 11 in SAS Data Science
‎03-12-2015 09:36 AM
‎03-12-2015 09:36 AM
Hi @msaber, I haven't used any of those BI tools. In the meantime, does it help to generate a pdf report using the Reporter node in EMiner? Did you get some input from your post in the Web Report Studio community? If not, try contacting SAS tech support using this form: Contact SAS Institute Technical Support thanks, Miguel ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎02-28-2018 11:39 AM
Quick links