Digital Risk Management: Credit Scoring Scorecard with Excel and Python	

In today's digital world, customers not only expect their shopping activity can be done online, they also demand their financial transactions also be done online and as quick as possible. Thus, financial institutions have to keep developing themselves following the latest developments in technology. 

For the credit lending division, it is crucial that they already moving from manual process and start implementing digital and automatic Credit Scoring, to determine potential users who are eligible for loans. Digitalization also unlocks new opportunity to utilize data that typically not used to assess loan eligibility. Thus, for traditional financial institutions to be able to survive in the digital environment, they need to know the framework of digital risk management and best practices in implementing Credit Scoring.

After completion of this program, participants will be better able to:
* Understand the concept of digital credit risk management and the use of Credit Scoring
* Identify appropriate data, as well as the use of alternative data, to be used to determine credit risk
* Formulate statistical framework to build Credit Scoring model
* Develop Credit Scoring with logistic regression using Excel and Python
* Transform the model into credit scorecard
* Implement best practices in digital Credit Scoring to determine loan approval

Certificate of completion (supplied by PasarTrainer)

Training Material

* Moving manual Excel calculation into Python coding
* Using scikit-learn to model Logistic Regression
* Evaluate Model performance using confusion matirx, K-S, and Gini
* Monitoring model with PSI and characteristic report


This module explains an overview of Visual Management


Module 3. Implemeting Credit Scoring in Python and how to evaluate model performance

A multiple-choice test to assess the participants' proficiency level prior to enrolling in the training program.


Pre-Test (10 Questions - Closed Book)

* The basic overview of Digital Risk Management
* The purpose of Credit Scoring within the context of Risk Management"


This module explains an overview of Digital Risk Management


Module 1. Overview of Digital Risk Management

* Each participant running the Python code and inform instructor of any issues


A group discussion where each group/participants are asked to share their Scorecard Model


Presentation : Group discussion on The Result of their Credit Scoring using Python 

* Group exercise : Each participant already prepare list of questions
* Debriefing : Instructor and other participants also answer the questions
* Instructor take note of their expectation and incorporate during the rest of the course


A group role play for each participant to share their experience and expectation from taking the course.


Case Study 

* How to create proper definition of bad customer
* Implementing more than one scorecard for loan approval
* Best practices in using data input and alternative data
* Choosing the right cut-off for loan approval


This module explains best practices in developing Credit Scoring and actual implementation of the Credit Scoring for loan approval


Module 4. Best Practices in Digital Risk Management

* Data preparation
* Exploratory Data Analysis
* Methods for data cleansing
* Data Engineering and Optimization of data input
* Modeling theory: Logistic Regression
* Create binning, Weight of Evidence, and Information Value
* Calculate Logistic Regression using Excel


This module explains an overview of statistic theory behind modeling used for Credit Scoring and how to manually calculate it using Excel


Module 2. Theory of Credit Scoring and Implementation in Excel

A multiple-choice test to assess the participants' proficiency level after enrolling in the training program.

* Group exercise : Credit Scoring calculation using Excel
* Presentation : present result of Credit Scoring


Each participants conduct exercise (case study) to practice on developing Credit Scoring
