Project Synopsis Cover Page

SYNOPSIS FOR TERM 2 PROJECT WORK
CREDIT RISK MODELING IN PYTHON
Submitted to Manipal Academy of Higher Education towards the partial

fulfilment of the Requirements for the Degree of Post Graduate Diploma in
Data Science
SUBMITTED BY
VINAYAK M RAO
19125760107
The student has to submit the following before he starts the Project for
approval
Project Synopsis of the proposed project work must contain –
 Project title
 The project synopsis should be at least of 3 pages
 Problem Definition/problem statement
 Introduction/Brief Description
 Scope & constraints (Project & Design)
 Tools used (Hardware/Software)
 Project Plan (Gantt Chart)
 Plan of Analysis/Performance Measurement
 Data Sets
 Data Sources
Problem Definition/problem statement
A creditor or lender provides goods and services to a borrower based on the trust that the
borrower will be able to repay the lender at some point in the future. In exchange for providing
these goods and services in the present, the lender receives a payment known as interest. The
likelihood that a borrower does not repay the loan amount plus the interest is known as credit
risk and the event of a borrower not being able to repay their debt is called default. In order to
protect themselves against default, lenders must assess credit risk associated with each
borrower very well. Credit risk is an important topic for the financial health of creditors
therefore the project will predict the probability of default models by abiding by the regulator
rules established in Basel 2 and Basel 3 as performed by banks.
Introduction/Brief Description
A creditor also known as a lender provides goods and services to a borrower based on the trust
that the borrower will be able to repay the lender at some point in the future. In exchange for
providing these goods and services in the present, the lender receives a payment known as
interest. Money is the most common type of loan provided by lenders to borrowers. Credit cards
and home loans are also example of credit provided by borrowers. A borrower can spend a
certain amount of money through their credit cards up till certain limit called credit limit. The
borrower repays this amount spent along with additional interest to the bank and this is how
the bank makes a profit. When we sign home ownership loans, we borrow money from the
bank against equity in the residence that we want to acquire and the residence covers the debt
in case we fail to repay. The bank has the right to take ownership of the property and sell it to
satisfy the remaining debt.
The likelihood that a borrower does not repay the loan amount plus the interest is known as
credit risk. The lender does not receive the principal amount and interest and also has to sustain
substantial costs to recover the outstanding debt called collection costs. This event of a
borrower not being able to repay their debt is called default. In order to protect themselves
against default, lenders must assess credit risk associated with each borrower very well.
Lenders inability to estimate the probability of default of borrowers leads to serious financial
crisis such as the one in 2008. The main factor that led to the crisis was the high default rate of
subprime mortgages. Lenders were willing to finance 100% of the value of a new home at low
interest rates which led to increase in demand and therefore increase in prices of these houses.
Homeowners borrowed money and used their houses as a guarantee to the bank but a lot of
them were high risk subprime and could not repay their loans and defaulted. Consequently,
financial instruments based on mortgages such as mortgage backed securities lost value and the
banks holding these instruments absorbed huge losses and some like Lehman Brothers went
bankrupt. This example shows why credit risk is one of the most important variables in the
financial system today.
Scope & constraints (Project & Design)

Credit risk is an important topic for the financial health of creditors therefore the project will
predict the probability of default models by abiding by the regulator rules established in Basel 2
and Basel 3 as performed by banks. Techniques such as coarse classing, fine classing, weight of
evidence and information value will be used to prepare the risk model . Once the data is set up,
3 models will be created-
1. Probability of default model which shows the probability of a customer not
repaying their debt.
2. Loss given default model showing the proportion of exposure a company loses
when a customer defaults.
3. Exposure at default model showing the total loss in terms of amount of money
the bank is exposed to if a customer defaults.
With the information from these three models, a scorecard is created which bank officers use to
accept or reject daily applications. The total expected loss for the whole portfolio of loans that
the bank holds is calculated by combining these three models. The expected loss is a key metric
for management.
Tools used (Hardware/Software)
1. Python- ver. 3.7.4

2. Tableau –ver.2019.4
Project Plan (Gantt Chart)

4/16/20204/26/2020 5/6/2020 5/16/20205/26/2020 6/5/2020 6/15/20206/25/2020
Preliminary data cleaning and EDA
Baseline Machine Learning Model
Feature selection & regularization
Mid Term Report preparation
Hyper Parameter Tuning and Insight generation
Further improvements
Final Report Preparation
Plan of Analysis/Performance Measurement (please add points)

Data Set
Lending Club Loan Data 2007-11 summary
These files contain complete loan data for all loans issued through the time period stated,
including the current loan status (Current, Late, Fully Paid, etc.) and latest payment information.
These files contain complete loan data for all loans issued through the 2007-2011, including the
current loan status (Current, Late, Fully Paid, etc.) and latest payment information. Additional
features include credit scores, number of finance inquiries, address including zip codes, and
state, and collections among others. The file is a matrix of about 42539 observations and 75
variables.
Data Sources https://www.lendingclub.com/info/download-data.action

Project Synopsis Cover Page

Uploaded by

Copyright:

Available Formats

Project Synopsis Cover Page

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Project Synopsis Cover Page

Uploaded by

Copyright:

Available Formats

SYNOPSIS FOR TERM 2 PROJECT WORK

CREDIT RISK MODELING IN PYTHON

Submitted to Manipal Academy of Higher Education towards the partial

Project Synopsis of the proposed project work must contain –

 The project synopsis should be at least of 3 pages

 Problem Definition/problem statement

 Scope & constraints (Project & Design)

 Tools used (Hardware/Software)

 Project Plan (Gantt Chart)

 Plan of Analysis/Performance Measurement

Scope & constraints (Project & Design)

Tools used (Hardware/Software)

1. Python- ver. 3.7.4

Project Plan (Gantt Chart)

Plan of Analysis/Performance Measurement (please add points)

You might also like