A CASE STUDY OF PERSONAL BANKRUPTCY PREDICTION USING DECISION TREE

Universidad de Sta. Isabel

Naga city
SY: 2020-2021

PERSONAL BANKRUPTCY PREDICTION USING DECISION TREE

A Case study
Presented to
Mrs. Lorna Icardi

By
Bengado, Regina S.
A CASE STUDY OF PERSONAL BANKRUPTCY PREDICTION USING DECISION

TREE INTRODUCTION
 Bankruptcy is a debt solution and a form of insolvency. It’s a legal procedure mainly

suited to people whose circumstances are unlikely to change and who have little hope of

paying off their debts within a reasonable time. Each one of us may experience personal

bankruptcy especially in the field of business, this issue cannot be avoided but can be

minimize if a person have a proper knowledge on how handling these kind of problems in the

business field. Personal bankruptcy is on the rise in Malaysia. The insolvency Department of

Malaysia

reported that personal bankruptcy has increased since 2007 from 13,238 cases to 22,351 cases.

This is indeed an alarming issue because the increasing number of personal bankruptcy cases

will have a negative impact on the Malaysian economy, as well as on the society. From the

aspect of individual’s personal economy, bankruptcy minimizes their chances of securing a

job.

By developing the personal bankruptcy prediction model using the decision tree

technique, the percentage of chance to bankruptcy may be predicted and minimize. A decision

tree is a flowchart-like structure in which each internal node represents a "test" on an attribute

(e.g. whether a coin flip comes up heads or tails), each branch represents the outcome of the

test, and each leaf node represents a class label (decision taken after computing all attributes).

Decision trees provide an effective method of Decision Making because they: Clearly lay out

the problem so that all options can be challenged. Allow us to analyze fully the possible

consequences of a decision. Provide a framework to quantify the values of outcomes and the

probabilities of achieving them. These actions are important to avoid or to prevent default

payment, bad debts and personal bankruptcy. Therefore, the objectives of this paper are to

identify the significant predictors and to determine the conditional probability of a borrower

belonging to a class using decision tree model. This study succeeds to come out with the

A CASE STUDY OF PERSONAL BANKRUPTCY PREDICTION USING DECISION TREE

profiles of bankrupts, reliable personal bankruptcy scoring model and significant variables or
personal bankruptcy.

BACKGROUND OF THE CASE

Personal bankruptcy issues have been a pressing concern to the governments, bankers,

creditors and financial researchers in recent years. Erroneous decision-making by financial

institutions will likely lead to financial crises and distress. Therefore, bankruptcy prediction

and credit scoring are very important in making financial decisions (Tsai, 2014). Jullamon

(2012) revealed that the bankrupts are usually those in their fifties, low-income earners,

unemployed, without property ownership and defaulted on the loan agreement. Livshits et al.

(2010) reported that credit market innovations or changes that reduce the cost of bankruptcy

(such as filing fees), together with the cost of borrowing (such as interest rates) play an

essential role toward the rise in bankruptcy cases. And according to Blan et al. (2007) the

causes of bankruptcy includes overspending on credit cards, poor financial management,

problems with house payment or property taxes and loss of jobs.

Despite the continual growth of personal bankruptcy cases since 2007, this issue

remains an under-investigated research area. Selvanathan et al. (2016) explored that factors

affecting personal bankruptcy cases based on a sample of Klang Valley residents using

Pearson correlation coefficient and multiple regression analysis. They found that there were

positive relationship between money management, financial literacy and non-performing loan

with personal bankruptcy.

Apart from the development of the bankruptcy prediction model, the accuracy of

personal bankruptcy prediction is a major issue to the shareholders, creditors, policy makers

and business managers. Credit scoring has been regarded as a core appraisal tool by different

A CASE STUDY OF PERSONAL BANKRUPTCY PREDICTION USING DECISION TREE

institutions for the last few years and has been widely investigated in different areas, such as
finance and accounting (Abdou and Pointon 2011). Decision tree (DT) is also commonly used

in data mining. It is frequently used in the segmentation of population or predictive models. It

is also a white box model that indicates the rules in a simple logic. Because of the case

interpretation, it is extremely popular in assisting users to understand various aspects of their

data.

The results showed that DT is the most accurate in financial distress prediction

provides an early warning signal and identify areas of weaknesses. Accurate corporate default

prediction usually leads to numerous benefits, such as cost reduction in credit analysis, better

monitoring and an increased debt collection rate. Hence, they used DT and LR technique to

develop a corporate default prediction model.

EVALUATION OF THE CASE

This study involved a data set obtained from an authorized debt management agency.

The data consisted of settled members and terminated members. Settled members were those

who managed to settle their loans, while terminated were those who were unable to pay their

loans. A DT model consists of a set of rules for dividing a large heterogeneous population into

smaller, more homogenous groups with respect to a particular target variable. The target

variable is usually categorical, and the DT model is used either to calculate the probability that

a given record belongs to each of the categories or to classify the records by assigning it to the

most likely class.

A CASE STUDY OF PERSONAL BANKRUPTCY PREDICTION USING DECISION

TREE PROPOSED SOLUTION

Bankruptcy includes overspending on credit cards, poor financial management,

problems with house payment or property taxes and loss of jobs.

1. All banks, entrepreneur, creditor and debtor must know how to make financial
 planning.

2. Each person must have at least 5 cards with different purposes. For example, card

for insurance, education, health and etc.

3. Avoid overspending on credit cards.

SYNTHESIS

This paper discussed the improvements in the classification of personal bankruptcy

using random understanding to correct the imbalanced data. The application of DT in this

study showed that the specificity rate had increased after the random understanding strategy

was applied.

1. This decision tree model is possible for patent and income generation. Financial

institutions are able to use this model for potential borrowers to predict their tendency

toward personal bankruptcy.

2. Create awareness to society on significant variables of personal bankruptcy so that they

can avoid being a bankrupt.

A CASE STUDY OF PERSONAL BANKRUPTCY PREDICTION USING DECISION

TREE Reference

https://www.google.com/search?ei=99V3X_XuL8GmmAXRlIWgBQ&q=decision+tree&oq=decision+tree&gs

_lcp=CgZwc3ktYWIQAzIFCAAQsQMyCAgAELEDEIMBMggIABCxAxCDATICCAAyAggAMgIIADICCA

AyAggAMgIIADICCAA6BQgAEJECOggILhCxAxCDAToLCC4QsQMQxwEQowI6BQguELEDOgoIABCR

AhBGEPkBOgsIABCxAxCDARCRAjoKCAAQsQMQgwEQQ1CmCliaGmDLHmgAcAB4AIABeogBtAmSA

QQxMC4zmAEAoAEBqgEHZ3dzLXdpesABAQ&sclient=psy

ab&ved=0ahUKEwi1xpDlpJfsAhVBE6YKHVFKAVQQ4dUDCA0&uact=5

https://www.google.com/search?source=hp&ei=8tV3X8vyJc

m0wTD37XIDA&q=perosnal+bankruptcy&oq=perosnal+bankruptcy&gs_lcp=CgZwc3ktYWIQAzIECAAQD

T
IECAAQDTIECAAQDTIECAAQDTIECAAQDTIECAAQDTIECAAQDTIECAAQDTIECAAQDTIECAAQ

DToICAAQsQMQgwE6BQgAELEDOgsILhCxAxDHARCjAjoCCAA6AgguOgUILhCxAzoICC4QsQMQgw

6BAgAEAo6BAguEAo6CgguEMcBEKMCEApQ3gRY2iFgySRoAHAAeACAAYABiAHdDZIBBDE2LjOY

AQCgAQGqAQdnd3Mtd2l6&sclient=psy

ab&ved=0ahUKEwjLs9XipJfsAhVP05QKHcNvDckQ4dUDCAc&uact=5

https://www.google.com/search?q=reason+of+bankruptcy&source=lnms&tbm=isch&sa=X&ved=2ah

UKEwiFksnrtJfsAhWGwZQKHSb4Cm8Q_AUoAnoECBAQBA&biw=1366&bih=662#imgrc=ZGV

bQZK6DSuhtM