All Discussions Tagged 'Credit' - AnalyticBridge2021-06-21T15:14:55Zhttps://www.analyticbridge.datasciencecentral.com/forum/topic/listForTag?tag=Credit&feed=yes&xn_auth=noCredit Risk Scorecards Points Calculationtag:www.analyticbridge.datasciencecentral.com,2013-07-13:2004291:Topic:2558432013-07-13T12:01:26.611ZRockyRambohttps://www.analyticbridge.datasciencecentral.com/profile/VarunNakra
<p>We use linear scaling to calculate score/points for an applicant. Now, if we were to determine the number of points for each attribute of a characteristic, Naeem Siddqui, gives the following equation (Image attached).</p>
<p>I understand that it is nothing but dividing the total # of points into each attribute of each characteristic which is why we are dividing the offset and the intercept term by number of characteristics, however, what perplexes me is the negative term. <br></br> ln(odds) has…</p>
<p>We use linear scaling to calculate score/points for an applicant. Now, if we were to determine the number of points for each attribute of a characteristic, Naeem Siddqui, gives the following equation (Image attached).</p>
<p>I understand that it is nothing but dividing the total # of points into each attribute of each characteristic which is why we are dividing the offset and the intercept term by number of characteristics, however, what perplexes me is the negative term. <br/> ln(odds) has been replaced by the logit equation with a negative term outside the bracket. I am unable to understand from where did the negative sign appear.<br/>
If anyone has an idea then please advise.</p>
<p>Thanks</p>
<p></p> Using a Hurdle Model in Credit Scoringtag:www.analyticbridge.datasciencecentral.com,2013-06-20:2004291:Topic:2522582013-06-20T18:53:10.144ZMichael Hhttps://www.analyticbridge.datasciencecentral.com/profile/MichaelHartye
<p>Has anyone used or have thoughts on using a 2-step hurdle model to address the imbalance of "GOODS" vs "BADS" often present in a sample of borrowers?</p>
<p>That is, first run a logistic regression on your Good vs Bad, then take all of your Bads and use the % paid on the loan as the dependent variable and run a separate linear regression. In the Linear, those who defaulted after 3 months would fair worst that those who nearly paid off completely.</p>
<p>Finally combine outcomes of the two…</p>
<p>Has anyone used or have thoughts on using a 2-step hurdle model to address the imbalance of "GOODS" vs "BADS" often present in a sample of borrowers?</p>
<p>That is, first run a logistic regression on your Good vs Bad, then take all of your Bads and use the % paid on the loan as the dependent variable and run a separate linear regression. In the Linear, those who defaulted after 3 months would fair worst that those who nearly paid off completely.</p>
<p>Finally combine outcomes of the two models to create a scorecard.</p>
<p>Any thoughts?</p>
<p></p>
<p>Thanks,</p>
<p>Mike</p> Macro Data in Credit Scorecard Modelstag:www.analyticbridge.datasciencecentral.com,2013-05-10:2004291:Topic:2450962013-05-10T18:06:28.287ZMichael Hhttps://www.analyticbridge.datasciencecentral.com/profile/MichaelHartye
<p>I was wondering if anyone has found any success in using external macro data (i.e. per capita income, unemployment rate, foreclosure data) in building out credit models. For the record, I model for small business loans where the personal credit data of the owner along with verifiable business information is typically most relevant to predicted performance.</p>
<p>I was wondering if anyone has found any success in using external macro data (i.e. per capita income, unemployment rate, foreclosure data) in building out credit models. For the record, I model for small business loans where the personal credit data of the owner along with verifiable business information is typically most relevant to predicted performance.</p>