Logistic Regression (Credit Scoring) Modeling using SAS

Analytics / Machine Learning / Data Science: Statistical / Econometrics foundation, SAS Program details, Modeling demo
4.7 (111 ratings) Instead of using a simple lifetime average, Udemy calculates a
course's star rating by considering a number of different factors
such as the number of ratings, the age of ratings, and the
likelihood of fraudulent ratings.
609 students enrolled
$70
Take This Course
  • Lectures 92
  • Length 16.5 hours
  • Skill Level All Levels
  • Languages English
  • Includes Lifetime access
    30 day money back guarantee!
    Available on iOS and Android
    Certificate of Completion
Wishlisted Wishlist

How taking a course works

Discover

Find online courses made by experts from around the world.

Learn

Take your courses with you and learn anywhere, anytime.

Master

Learn and practice real-world skills and achieve your goals.

About This Course

Published 5/2014 English

Course Description

What is this course all about?

This course is all about credit scoring / logistic regression model building using SAS. It explains

There course promises to explain concepts in a crystal clear manner. It goes through the practical issue faced by analyst. Some of the discussion item would be

  • How to clarify objective and ensure data sufficiency?
  • How do you decide the performance window?
  • How do you perform data treatment
  • How to go for variable selection? How to deal with numeric variables and character variables?
  • How do you treat multi collinerity scientifically?
  • How do you understand the strength of your model?
  • How do you validate your model?
  • How do you interpret SAS output and develop next SAS code accordingly?
  • Step by step workout - model development on an example data set

What kind of material is included?

It consists of video recording of screen (audio visual screen capture), pdf of presentations, Excel data for workout, word document containing code and Excel document containing step by step model development workout details

How long the course will take to complete?

Approximately 30 hours

How is the course structured?

It has seven sections, which step by step explains model development

Why Take this course?

The course is more intended towards students / analytics professionals to

  • Get crystal clear understanding
  • Get jobs in this kind of work by clearing interview with confidence
  • Be successful at their statistical or analytical profession due to the quality output they produce

What are the requirements?

  • Basic knowledge of SAS

What am I going to get from this course?

  • Learn model development
  • Understand the science behind model development
  • Understand the SAS program required for various steps
  • Get comfortable with interpretation of SAS program output
  • See the step by step model development

What is the target audience?

  • Students
  • Analysts / Analytics professional
  • Modelers / Statisticians

What you get with this course?

Not for you? No problem.
30 day money back guarantee.

Forever yours.
Lifetime access.

Learn on the go.
Desktop, iOS and Android.

Get rewarded.
Certificate of completion.

Curriculum

Section 1: Course Outline
Course content
Preview
06:31
Introduction to logistic Regression Modelling - High level
Preview
08:34
Udemy Content details - Model workout details and excel file downloads
Preview
03:32
Tips for Students
02:55
Course Content PDF
3 pages
Section 2: Introduction to Credit Scoring / Credit Score card development
Section outline
Preview
01:23
3C Concept of Credit Approval Process
15:31
High Level Understanding of Score
09:31
Benefit of scoring (modelling)
20:25
Introduction to modeling
07:09
Types of scores
Preview
12:47
A typical risk score
04:16
Section PDF
20 pages
Section 3: Data Design for Modelling
Section outline
Preview
02:44
Model Design Example
17:54
Model Design - definitions and pointers
13:19
Decide Performance window by Vintage Analysis
Preview
14:51
Model Design Precaution
08:18
Section PDF
20 pages
Section 4: Data Audit - Make sure to check that data is right for the modelling
Section Outline
Preview
03:59
Essential Data Quality
03:45
Getting free access to SAS
10 pages
How to download excel / word files ?
Preview
02:39
09:02

Download Excel and word file for the explanation provided as part of feel the data section

Feel the data - View it's contents
09:29
Feel the data - know it's distinct values
09:02
Feel the data - know it's distribution
13:27
Feel the data - Understand Coefficient of variance (need and applicability)
08:16
Feel the data - know kurtosis and skewness
04:47
Feel the data - know the percentile
11:21
Feel the data - know stem n leaf diagram
Preview
05:35
Feel the data - Understand box plot to detect outliers
06:15
Feel the data - Understand and interpret normal probability plot
22:27
Missing Value treatment And Flooring / Capping Guidiline
13:49
Section PDF
31 pages
Section 5: Variable Selection - Select important numeric and character variables
Section Outline
Preview
03:00
Variable Selection - High level and flow chart of steps
13:04
Important Character / Categorical Variable selection - high level
Preview
06:07
19:52

Please download the excel for better understanding!

Getting Chi-Square statistics using SAS
08:36
11:02

Please download two excel and the word document for model / data work out

Model Workout - 01 Data Treatment
34:52
Numeric Variable Selection - Part 01
10:43
SAS Macro to check directional sense of numeric variable
14:58
Recap Linear Regression
04:08
Introduction to Logistic Regression
11:57
Theory and Example of Step wise selection of Numeric Variable
19:53
Appendix - Fisher's linear discriminant function to select important numeric Var
09:23
Appendix - Information Value method of selecting important variables (all types)
10:07
Appendix -Phi Square and Cramer's V for important categorical variable selection
06:59
Section PDF
64 pages
Section 6: Multi Collinearity Treatment
Section Outline
Preview
03:12
Common Sense Understanding of Multi collinearity and it's impact
07:02
Detecting Multi Collinearity
10:10
Multi Collinearity Treatment - part 01
19:20
Multi Collinearity Treatment - part 02
05:58
Model Data workout - 02 Bi Variate strength of variables
09:41
Model Data workout - 03 Multi Collinearity Treatment (Scientifically)
12:36
Section PDF
24 pages
Section 7: Iterate for final model / Understand strength of the model
Section Outline
Preview
03:05
Introduction to final model development steps
04:07
Logistic Model Information - part 01
05:53
Logistic Model Information - part 02
04:48
Model Fit Statistics
02:58
Log Likelihood
15:12
Log Likelihood ratio - part 01
06:28
Log Likelihood Ratio - part 02
03:29
Model Fit Statistics - Revisit
13:23
Maximum Likelihood Estimate
12:44
Concordance, Somer's D, Gamma, Tau etc.
Preview
18:47
Ideal logistic regression output
04:17
Model Data Workout - part 04 Try Model on 10 variables
06:50
Model Data Workout - part 05 Select best 8 variables
09:26
Section PDF
39 pages
Section 8: Strength of a Model and Model Validation Methods
Section Outline
Preview
02:21
Model Data Workout - part 06 Coefficient Stability Check
11:28
Understand Score and Generate Score in the data set
12:21
Theoretical Understanding of KS
Preview
03:59
Model Data Workout - part 08 Generate KS Statistics for the model
20:51
Model Data Workout - part 09 Understand and Generate Gini Statistics
11:30
Model Data Workout - part 10 Understand & Apply Model Validation n Stability Chk
07:56
Model Presentation Guideline - What should be presented to business
05:00
Final Words
01:59
Section PDF
25 pages
How to download excel / word files ?
2 pages
Section 9: Reject Inference - Developing application score on scored population
Section Overview
Preview
00:43
Introduction to reject inference! What it is? Why it is needed?
11:21
How to do reject inference?
08:40
Impact of the new model - swapset analysis / more base with same approval rate
Preview
08:03
Swapset analysis supplementary video
03:07
Do you need reject inference all the time?
03:49

Students Who Viewed This Course Also Viewed

  • Loading
  • Loading
  • Loading

Instructor Biography

Gopal Prasad Malakar, Credit Card Analytics Professional - Trains on Data Mining

I am a seasoned Analytics professional with 15+ years of professional experience. I have industry experience of impactful and actionable analytics. I am a keen trainer, who believes that training is all about making users understand the concepts. If students remain confused after the training, the training is useless. I ensure that after my training, students (or partcipants) are crystal clear on how to use the learning in their business scenarios. My expertise is in Credit Card Business, Scoring (econometrics based model development), score management, loss forecasting and MS access based database application development.

Ready to start learning?
Take This Course