https://communities.sas.com/t5/New-SAS-User/How-to-process-multiple-responses-in-SAS-Miner/m-p/749227#M29557 <P>May I ask, in your mind, how do you want to handle companies with more than one credit facility? What would the logical steps you would use to take &gt;137000 records and condense them down into 37000 records? I am not asking you to say how to do it in SAS Enterprise Miner, I want you to explain in words what the steps would be.</P> Mon, 21 Jun 2021 10:41:27 GMT PaigeMiller 2021-06-21T10:41:27Z https://communities.sas.com/t5/New-SAS-User/How-to-process-multiple-responses-in-SAS-Miner/m-p/749223#M29556 <P>Hi,</P><P>&nbsp;</P><P>I am a new user for SAS Enterprise Miner.</P><P>&nbsp;</P><P>Currently, my plan is to develop model for bankruptcy prediction.</P><P>My dataset consist of:</P><UL><LI>1 target variable – binary – to predict an individual will be bankrupt (0) or non-bankrupt (1)</LI><LI>17 input variables – all measurement level is set to categorical (nominal)</LI></UL><P>&nbsp;</P><P>Sample of my dataset is compiled as follows in Excel. One client may have more than one type of credit facility. For example, client no. AQ4 has 5 types of credit facilities: study loan, mortgage, personal loan, credit card and hire purchase (loss after sales).</P><P><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="ANMS_0-1624270705039.png" style="width: 400px;"><img src="https://communities.sas.com/t5/image/serverpage/image-id/60551iBEE5A61EF0549674/image-size/medium?v=v2&amp;px=400" role="button" title="ANMS_0-1624270705039.png" alt="ANMS_0-1624270705039.png" /></span></P><P>&nbsp;</P><P>The dataset is compiled in Excel file. When it is being compiled like this, it generates 137,392 rows in Excel.</P><P>&nbsp;</P><P>Then, I import the dataset into SAS Miner to analyse the dataset. Suppose, I have about 37,000 DMP clients (30,000 bankrupts + 7,000 non-bankrupts).</P><P><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="ANMS_1-1624270705055.png" style="width: 400px;"><img src="https://communities.sas.com/t5/image/serverpage/image-id/60553iA0D2A42F894BC217/image-size/medium?v=v2&amp;px=400" role="button" title="ANMS_1-1624270705055.png" alt="ANMS_1-1624270705055.png" /></span></P><P>&nbsp;</P><P>However, the result given in the dataset provides the number of rows as observation: 137,392; which I think, is incorrect as my total clients (respondents) is 37,000 only. This happen most probably due to the type of credit facility (which 1 client has more than one facility type).</P><P><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="ANMS_2-1624270705058.png" style="width: 400px;"><img src="https://communities.sas.com/t5/image/serverpage/image-id/60552iF8213A2C5AE38696/image-size/medium?v=v2&amp;px=400" role="button" title="ANMS_2-1624270705058.png" alt="ANMS_2-1624270705058.png" /></span></P><P>&nbsp;</P><P><STRONG>Therefore, may I know how to compile / process this kind of variable with multiple responses using SAS Miner so that the output will generate the correct figures?</STRONG></P><P>&nbsp;</P><P>Appreciate if anyone could provide me detailed guide to solve this issue. Many thanks in advance.</P><P>&nbsp;</P><P>&nbsp;</P> Mon, 21 Jun 2021 10:19:31 GMT https://communities.sas.com/t5/New-SAS-User/How-to-process-multiple-responses-in-SAS-Miner/m-p/749223#M29556 ANMS 2021-06-21T10:19:31Z https://communities.sas.com/t5/New-SAS-User/How-to-process-multiple-responses-in-SAS-Miner/m-p/749227#M29557 <P>May I ask, in your mind, how do you want to handle companies with more than one credit facility? What would the logical steps you would use to take &gt;137000 records and condense them down into 37000 records? I am not asking you to say how to do it in SAS Enterprise Miner, I want you to explain in words what the steps would be.</P> Mon, 21 Jun 2021 10:41:27 GMT https://communities.sas.com/t5/New-SAS-User/How-to-process-multiple-responses-in-SAS-Miner/m-p/749227#M29557 PaigeMiller 2021-06-21T10:41:27Z