Credit model

[DEBOER]

Hi,
I'm going to start building credit risk models (scorecards), validation, etc.. A credit score is a numerical expression based on a statistical analysis of a person's credit files, to represent the creditworthiness of that person. Probability that a person pay back their loans. I need to practice a bit, is there someone who has access to a link, or file and document that describes the entire process. Everything I've looked at cost a lot of money and it is only theory they are going through. I can theoretically but I want to carry out the practical in SPSS. Thank you in advance

[Carlos Renato]

Dear friend

I worked with a risk analysis of credit here in Brazil. Everything starts with a strong data mining, evaluation of the consistency of information, building models (usually logistic regression or neural networks, some also use discriminant analysis and other techniques of optimization via genetic algorithms). Many companies use the methodology proposed by CRISP-DM with changes for each case.

Techniques such as CHAID, tree analysis, generation of rules of customer behavior are also common.

It is also common to compare new customers with customers used to model through some tests of compliance data.

What exactly did you search?

Carlos Renato
Statistician - Brazil

[DEBOER]

Yes exactly logistic model, I wonder if anyone has a description of how the process works? Step by step in SPSS

[Art Kendall]

I do not know of a file, but you could simulate one with
INPUT PROGRAM


Art Kendall
Social Research Consultants

On 11/24/2011 9:40 AM, DEBOER wrote:

Hi,
I'm going to start building credit risk models (scorecards), validation,
etc.. A credit score is a numerical expression based on a statistical
analysis of a person's credit files, to represent the creditworthiness of
that person. Probability that a person pay back their loans. I need to
practice a bit, is there someone who has access to a link, or file and
document that describes the entire process. Everything I've looked at cost a
lot of money and it is only theory they are going through. I can
theoretically but I want to carry out the practical in SPSS. Thank you in
advance

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Credit-model-tp5020338p5020338.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Carlos Renato]

Analyze>Regression>Binary Logistic

  GOOD or BAD consumer is the dependent and others variables are independents.
There must be some document that deals with the SPSS logistic binomial model.

Carlos Renato

[Bruce Weaver]

After Analyze>Regression>Binary Logistic, click on Help in the dialog that appears, then click on Show Me on the page that opens.

You could also look at David Garson's "Statnotes" and the "textbook examples" from the UCLA computing website.  Use Google to find both.

HTH.

Carlos Renato wrote

Analyze>Regression>Binary Logistic

  GOOD or BAD consumer is the dependent and others variables are independents.
There must be some document that deals with the SPSS logistic binomial model.

Carlos Renato