Professional Documents
Culture Documents
4, October 2010
ISSN: 2010-0248
Risk Classification
9. INT
RO
DU
CTI
ON
1. Business
Environment
India is witnessing a
rapid upsurge on the
economic front. Although,
it had to battle with the
storm of global financial
crisis recently, it survived
and has so far, surpassed
most of its competitors. It is
still facing the challenges of
the grave economic crisis
and is on the sprint to
recover its former growth
rate.
The strong recovery of
Asia's
third
largest
economy in the last 10
months by successfully
overcoming the global
meltdown is proof of the
country's
power
of
resilience. The projected
GDP growth rate is 8.4 per
cent in 2010 - 11.
A developing economy
like India always craves for financial resources. NBFCs in India: NonDemand for credit is great and often organized banking
Financial
traditional financing institutions (like banks andCompanies (NBFCs) play a
financial institutions) do not meet such demand thus vital role in the context of
creating a space for other types of financing. Money Indian Economy. They are
lender is an age old institution filling such space. indispensable part in the
Opening up of economy gave a further boost to the Indian financial system
demand for credit.
because they supplement
At this juncture, NBFCs, which basically were the activities of banks in
better organized money lenders happened in large terms
of
deposit
number. NBFCs in India have played a useful role in mobilization and lending.
financing various sectors of the economy, particularlyThey play a very important
those that have been underserved by the banks. In fact, role by providing finance to
many banks are forming NBFCs to take advantage of activities which are not
their greater flexibility in dealing with customers.
served by the organized
362
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
a
r
l
i
e
r
C. Key Definitions
C
r
Credit Risk: Credit risk arises
e
from the failure of counter party to
d
fulfill its contractual obligations
i
either during the course of a
t
transaction or on a future
obligation. Default is a discrete
S
state for the counter party. That is,
c
at best; the counter party makes
o
the requisite payment resulting
r
into a no loss situation and at
e
worst, the entire sum due is lost. A credit score is a
As such, it is a potential loss of numerical expression
valuable assets due to the probable based on a statistical
deterioration
in
the
credit analysis of a person's
worthiness of the counter party or credit files, to represent
if the counter party is not in a the creditworthiness of
position to fulfill its contractual that person. A credit
obligations. As such, credit risk score
is primarily
would inevitably follow underbased on credit report
such a scenario.
information typically
Credit Event: Credit eventsourced from credit
refers to a default resulting into abureaus. Lenders, such
no payment situation on anyas banks and credit
financial obligation. ISDA Mastercard companies, use
Netting Agreement for creditcredit
scores
to
derivatives has categorized theevaluate the potential
following events as credit event. risk posed by lending
Bankruptcy
money to consumers
1 Failure to make payment due and to mitigate losses
2 Obligation/cross
defaultdue to bad debt.
Scoring:
resulting from occurrence of a Credit
default on any other similar Credit scoring is an
obligation not being a failure eminently pragmatic
and empirical approach
to make payment due
3 Restructuring resulting intoto the problem of risk
of
terms that are less favorable assessment
individual debtors. The
t
h
emphasis
is
on
a
predicting defaults, not
n
on explaining why they
do or dont occur. The
e
363
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
models.
Thus,
the
models appear to be
more generally useful
for predicting financial
distress,
not
just
bankruptcy. In sum, the
results of this study
suggest that researchers
should use bankruptcy
prediction
models
cautiously.
Applying
the models to time
periods and industries
other than those used to
develop the models
may result
in
a
significant decline in
the models' accuracies.
Additionally,
some
bankruptcy prediction
models may be more
appropriate
for
evaluating
various
forms
of financial
distress as opposed to
just bankruptcy. To
avoid
erroneous
applications
of
bankruptcy prediction
models in the future, it
is
necessary
for
researchers not only to
understand the uses of
prediction models, but
also to understand the
limitations
of
the
models.
Mark
Schreiner
(2000), Credit Scoring
for Microfinance: Can
it work?, In rich
countries, lenders often
rely on credit scoring
formulae to predict
risk based on the
364
performance of past
loans
with
characteristics similar
to current loansto
inform decisions. Can
credit scoring do the
same for microfinance
lenders
in
poor
countries? This paper
argues that scoring
does have a place in
microfinance. Although
scoring is less powerful
in poor countries than
in rich countries, and
although scoring will
not replace the personal
knowledge of character
of loan officers or of
loan groups, scoring
can improve estimates
of risk. Thus, scoring
complementsbut
does not replace
current
microfinance
technologies.
Furthermore,
the
derivation
of
the
scoring formula reveals
how the characteristics
of borrowers, loans,
and lenders affect risk,
and this knowledge is
useful whether or not a
lender uses predictions
from scoring to inform
daily decisions. In the
next decade, many of
the
biggest
microfinance lenders
will likely make creditscoring models one of
their most important
decision tools.
Montserrat Guillen
and Jose M.
Martinez (1994), A
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
combines
it
with
Model for Credit Scoring An subjective data about a
Application of Discriminant company
and
the
Analysis, The application of environment in which
statistical techniques in decision it operates. As the new
making, and more specifically for credit scoring model
classification requirements, has combines factual and
proved to be adequate in the subjective information,
context of financial problems. it can be used to
This
paper
presents
the evaluate all aspects of
methodology used and the results the company (lessee)
obtained in the elaboration of a and draw an overall
decision-support system for credit assessment of viability
assignment. The paper explains and creditworthiness.
the statistical techniques and the This
model
is
characteristics of data used for applicable only for
eliminating
discriminating auto ancillary sector
function. Some comments about Objective:
The
the application of the model are primary objective of
given and results concerning the this research is to
optimal level of risk are also provide a robust credit
discussed, in order to give clear scoring model which is
patterns for implementation.
free from subjectivity
Yong H. Kim and Venkatand has got empirical
Srinivasan
(1987),
Aproof for decision
Comparative
Analysis
of making. In this study
Classification
Procedures,we made an attempt on
financial classification issues, anddesign and develop a
particularly the financial distresscredit scoring model
problem, continue to be subject towhich is specific for
vigorous
investigation.
TheIndian Auto Ancillary
corporate credit granting processSector.
has not received as much attention Research Design
in the literature. The corporate Nature
of
the
credit granting process has notStudy: The study is
received as much attention in thedescriptive.
A
literature. This paper examines thedescriptive study is
relative
effectiveness
ofundertaken in order to
parametric, nonparametric andascertain and be able
judgmental
classificationto
describe
the
procedures on a sample ofcharacteristics of the
corporate credit data. Thevariables of interest in
judgmental model is based on thea situation.
Analytic
Hierarchy
Process. Sampling Design
Evidence
indicates
that Sources of Data:
(nonparametric)
recursiveThe data used is of
partitioning methods providesecondary type. The
greater
information
thandata were collected
simultaneous
partitioningfrom the applicants
procedures. The judgmental modelwebsites and annual
is found to perform as well asreports. The existing
with
the
statistical
models.
Arecords
company
were
also
complementary relationship is
as
the
proposed between the statisticalconsidered
source
of
data.
and the judgmental models as an
effective paradigm for granting Population: As the
model is built mainly
credit.
to classify the lease
applicants from Auto
III. RESEARCH
Ancillary Industry, the
METHODOLOGY
Need for the Study: The new population represents
model takes data from the all the auto ancillary
financial
components
and units in India.
quantitative
and
qualitative, and the
proposal of the research
team
have
been
discussed in a top level
committee.
Schematically,
the
process is presented in
figure 4.1. It
is
interesting to see all the
things
that
are
considered, and the
hierarchy assigned to
each, in order to arrive
at a rating and that
qualitative analysis is
placed at the very top
of the pyramid.
Figure 4.1 Standard
Considerations in the Rating
Process
Source:
www.v3.moodys.com
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
from a downturn or
B. Credit Scoring
recession.
Also
Methodology for
industries,
being
cyclical
in
nature
Manufacturing
appear to suffer more
Companies
A good scorecard should severely from adverse
consider a number of financial and economic
non financial parameters of the shocks/conditions.
competition
enterprise and the impact of the Further
risk
coupled
with
macro economic factors like
technological
level
of
government policies, trade polices
competition
in
the
and regulations and the industry
industry would define
specific dynamics.
The industry in which a the bargaining power
company operates has a direct of its players, the
bearing
on
the
overall intensity of margin
and
the
performance of the company and pressures
of
therefore, the company should be degree
rated
based
on
industry competitiveness
required
for
the
benchmarks.
Further the dynamics or the risk business to exist in the
run.
Further
factors affecting a small company,long
a medium size job manufacturer or Government policies
a large manufacturer are different play a major role in
the
and hence the size of the firm determining
outlook
of
the
industry.
should be considered.
The rating framework includes the D. Business Risk
following risks:
Business risk is the
1 Industry Risk
possibility of a credit
2 Business Risk
customers failing to
3 Management Risk
pay
because
of
4 Financial Risk
circumstances
connected with the
C. Industry Risk
customers
business
The industry in which an
activities
and
enterprise operates plays a crucial
management. Business
role in the credit risk assessment.
risk can impact a
It is a key determinant of the level
company
at
the
and volatility in earnings of any
enterprise,
business
business. Other factors remaining
unit
or
business
the same, industry risk determines
process level. The
the ratings. Some of the factors
elements of business
that are analyzed include:
risks can be identified
1 Demand Factors
under the broad heads
2 State of Competition
of market risk and
3 Environmental Factors
operational efficiency
4 Bargaining Power of suppliers risk.
The industry risk also considers 1) Market Risk
economic risk arising from Market risk is the
economic
instability,
and exposure of the unit
depressed
or
deteriorating to the forward and
economic conditions within a
country. Decline in a countrys
economy, or in one of its
particular industries, should arise
concerns about whether to grant
credit to customers in that sector,
or if so, whether to limit the credit
exposure to that sector/ industry.
The risks of industry exposure are
likely to be greatest when the
domestic economy is suffering
backward linkage in
the
course
of
conducting
its
business, and the risk
of facing sustained
periods of unfavorable
trends in such factors
as product prices, raw
material prices, single
product
dependence,
pricing
inflexibility,
etc.
The
important
factors
should
be
considered
for
assessing market risk
are:
1 Positioning of the
products (Luxury
Vs Necessity)
2 Distribution
Network (Owned
or Rented)
3 Relationship with
Customers ( Fixed
orders, Number of
Customers)
4 Product
Range
(Single
Vs
Multiple)
2) Operational
Efficiency Risk
In markets where
competitiveness
is
largely determined by
costs,
the
market
position is determined
by
the
units
operational efficiency.
The result of these
factors is reflected in
the ability of the unit to
maintain /improve its
market
share
and
command differential
in pricing. In a
competitive market, it
is critical for any
business unit to control
its costs at all levels.
This assumes greater
importance
in
commodity or "me too
" businesses, where
low cost producers
almost always have an
edge.
Cost
of
production to a large
extent is influenced by
Location
of
the
production
unit(s),
Access
to
raw
result in optimum
utilization of resources.
I. Technology
Technological risk is
the threat that new
product / service or
inventions poses a
challenge to existing
product
/
service
whereby demand for
existing product will
either reduce or be
eliminated.
Two
primary sources of
technological risk are
rate of obsolescence
and
difficulties
associate with adoption
which
primarily
involves higher cost,
and
may
face
incompatibility
problem.
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
better understanding of
J. Management Risk
the influence of all the
Management risk refers to the business and financial
instance of risk of non payment risk factors. Evaluation
arising out of a business failure of the existing financial
is
also
due to the perceived inefficacies position
for
of the management. The elements important
the
in management risk are assessing determining
the management quality judged on sources of secondary
the basis of the basic educational cash flows and claims
qualification,
professional that may have to be
experience of the entrepreneur; serviced in future.
and business attitude that is related L. Indicators of
to the motivation of carrying out Financial
the business and pursuing business Performance:
strategies.
Financial
Character relate to theindicators over the last
willingness to pay. Apart from thethree
years
are
characteristic
disposition
of analyzed
and
honesty and integrity, several performance of the
aspects are judge in terms of
enterprise is compared
1. Track record of previouswith
its
peers.
borrowing and payment is an Comparison with peers
indicator.
is important for better
2. Whether
the
owners/ understanding of the
directors have a financial interest industry trends and
determining
the
in the business.
Ability relates basically the relative position of the
ability to pay. Credit worthiness ofissuer. Some of the
the entity is assessed, including its important
indicators
financial strength.
that are analyzed are
Capacity refers to thepresented below:
borrower
having
technical,1 Gross
profit
managerial and financial abilities
margin
in order to operate profitably and2 Operating profit
succeed in business.
3 Net profit margin
Past
experience
of
the
4 Return on capital
management in handling similar
employed
business, performance of group
5 Return on net
companies and their track record,
worth
vision and mission of the
6
Total debt as a %
management,
organization
of tangible net
structure, succession issues, net
worth
worth and corporate governance
7
Long term debt as
also plays an important role in
a % of tangible net
assessing the management.
worth
K. Financial Risk
8 Total
outside
Financial
risk
analysis
liabilities as a % of
tangible
total
involves thorough evaluation of
assets
the financials of the companies.
Careful analysis of the audited9 Interest coverage
ratio
financials,
observations
of
auditors in the auditors report and
notes to accounts, consistent
treatment of financials play an
important role.
While the focus of rating
exercise is to be determined the
future cash flow adequacy for
servicing debt obligations, a
detailed review of the past
financial statements is critical for
1 Debt
2
3
service
coverage ratio
Current Ratio
Quick Ratio
M. Developing a
Credit Scoring
Model Specific for
Auto Ancillary
Industry
As the model was
developed to classify the
applicants from Auto
Ancillary Industry, the
risk factors applicable
only to the auto ancillary
units were taken into
consideration.
There
were
28
factors in total, taken
for developing the
model.
The
list
includes
both
the
qualitative
and
quantitative factors.
The factors were:
1 Years of existence
of the company
2 Location of the
company
3 Group to which
the
company
belongs
4 Distribution
network
5 Relationship with
customers
6 Product range
7 Location of the
production units
8 Access to human
resources
9 Technology
10 Track record of
management
11 Operating profit
12 Share capital
13 Current ratio
14 Quick ratio
15 D/E ratio
16 Interest coverage
ratio
17 Inventory turnover
ratio
18 Debtors turnover
ratio
19 Fixed
assets
turnover ratio
20 Working capital
turnover ratio
21 Return on capital
employed
Chennai,
Bangalore,
Mumbai, Hyderabad,
22 PAT/Net sales
Pune,
Delhi,
and
23 PAT as % of Total Assets
Kolkata.
This
is
mainly
24 PAT/Net worth
to ensure a better
25 Growth in Sales
quality of service and
26 Growth in Profitability
ease in collection of
27 Net working capital
rentals.
However,
28 Working capital as % of total leases
are
also
asset
extended
at
other
Years of Existence of the centres to where there
company: The data value is the are no branches.
actual number of years of Values: 2 = The
existence.
NBFCSs comfortcompany is located in
levels are better with firms with favorable places, 1 =
longer establishment.
The Company is in
Location of the company:non-branch locations
NBFCS mainly
leases
to Group to which the
companies located in the cities company
belongs:
367
Values:
3
=
Company has had a
long term relationship
with NBFCS, 2 =
Company belongs to a
wellestablished
group, 1 = Company
does not belong to any
major group
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
performance.
Distribution network: A well Share
Capital:
established distribution network Authorized
share
which is owned by a company is a capital is the total
clear indicator of capacity of its amount
of
shares
business.
available to be issued.
Values: 2 = Rented distribution Issued share capital
network, 1 = Owned distribution relates to the total
network
amount
of
shares
Relationship with customers:actually subscribed for.
This
factor
measures
the Current Ratio: It is
companys relationship with itsthe ratio of the current
customers in terms of number ofassets to the current
customers, performance of theliabilities and measures
customers in the industry, andthe ability of the firm
number of fixed orders.
to meet its current
Values: 2 = Good customers liabilities. The higher
with fixed orders, 1 = No fixed the ratio, greater the
orders
short term liquidity. .
Product
range:
Risk Quick Ratio: Quick
associated with a company that ratio measures how
manufactures only one product is much of the short term
higher than that of a company that and current liabilities
manufactures multiples products. and provisions are
Values: 1 = Single product, 2 = covered by the highly
Multiple product
liquid current assets.
Location of the production Inventories are not
units: When a company has included
in
its
productions units which are close calculation as they are
to suppliers and markets, the deemed to be the least
operating efficiency of the liquid component of
company goes up.
current assets.
Values: 2 = Multiple and
Debt/Equity Ratio:
favorable locations, 1 =
This ratio is very
Unfavorable locations
important ratio to
Access to human resources:
judge effectiveness of
This factor measures
the long term financial
the employee turnover rate.
policy of the business.
Values: 2 = Low
It is the relationship
turnover rate, 1 = High between the external
turnover
and internal liabilities
rate
of a concern.
Technology:
This
factor
Interest Coverage
measures if the company is
Ratio:
This
ratio
adapting
to
the
recent
measures the number
advancements in technology.
of times the fixed
2 = New technologies used, 1 =
interest
charge
is
Old technologies used
covered by the periodic
Track record of management:
profit. It is always
Track record of management is a
desirable to have profit
qualitative factor and is crucial in
more than interest
building credit scoring model.
payable.
Track record of management
Inventory Turnover
includes the years of experience
Ratio: It establishes the
of the entrepreneur, education,
previous experience in the relationship between the
industry,
strength
of
core cost of goods sold
during the year and
management group, etc.
Values: 3 = Good, 2 = Average, average
1 = Poor Operating Income: It
is the top- line of the company.
Higher the value of operating
income, the better the firms
Return on Capital
Employed:
A
commonly used rate of
return measure, the
return
on
Capital
Employed (also called
as return on total assets
or
Return
on
Investment) is defined
as the ratio of Net
Income (profit) to
average Total Assets.
Classification of
Applicants
The
variables
discussed
in
the
previous section were
treated as independent
variables with Risk
Level being treated as
dependent variable and
a model was built.
The
dependent
variable Risk Level
could take three values.
This data is of nominal
type.
The values are:
1 = Low Risk, 2 =
Moderate Risk, 3
= High Risk
The
initial
classification
of
companies based on
the risk level was done
using existing data
with the company. Out
of
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
function.
variables
in
classifying
the
dependent
variable.
To
discard
variables
which
are little related to
group distinctions.
25 Growth in Sales
26 Growth
in
Profitability
working
capital
28 Working capital as
% of total asset
27 Net
V. DATA
ANALYSIS AND
INTERPRETATIO
N
Total number of
cases taken for the
analysis was 70. Out of
these 70 cases, 50 were
considered for building
the model and 20 were
used for validation.
Discriminant Analysis
was
performed
to
obtain a function that
discriminates
the
applicants
(Lessees)
into
three
riskcategories, low risk,
moderate risk, and high
risk.
The
values
of
independent
and
dependent variables of
50 cases taken for
building the model
were used as the inputs
and the discriminant
functions
were
obtained. As mentioned
earlier,
two
discriminant functions
would be derived as
there
were
three
groups. Later, the
values of another 20
cases were added to
validate the model.
A. Case Processing
Summary
Case Processing
Summary is to find
out if all the cases are
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
5.3
EIGE
N
VAL
UES
are
notis the ratio of
significant. the betweenFrom thegroups sum
table 5.2, itof squares to
can
bethe withininferred thatgroups sum
there
areof
squares.
strong groupThe largest
differences Eigen value
for the factorscorresponds
Group
toto
the
which
theeigenvector
company
in
the
belongs,
direction of
Relationship the maximum
with
spread of the
customers, groups
Location ofmeans.
the
The
productions
second
units, Product largest
range,
D/E Eigen
ratio, Return value
on
capital correspond
s to the
employed,
PAT/Net
sales,
and
Growth
in
profit.
Eigen
Values: The
Eigen value
B. WILKS
LAMBDA
Wilks'
lambda is the
proportion of
the
total
variance
in
the
discriminant
scores
not
explained by
differences
among
the
groups.
Wilks'
lambda
ranges
between
0
and 1. Values
close to 0
indicate the
group means
are different.
Values
close to 1
indicate the
group means
are
not
different
(equal to 1
indicates all
means are the
same).
not differ.
would
bemeasured in
TABLE
A
chi-square
more helpfuldifferent
5.4
transformation
of
units,
the
on
WILKS
Wilks' lambda is used
magnitude
of
classification
LAMBas
along
with
the
an
the
DA
degrees of freedom to
significance unstandardize
determine
value
ared coefficient
significance. If the
provides little
0.032.
significance value is
Standardi indication of
small (less than say
the relative
zed
0.10) this indicates
Discrimin contribution
that group means Interpreta
ant
of
the
differ.
If
thetion:
Function variable
Table
to
Coefficient the
significance value is5.4
shows
overall
s:
large (more than saythat the first
discriminatio
When
0.10) this indicatesdiscriminant
n.
variables are
that group means dofunction
370
Standardizing
the
coefficients
allows us to
examine the
relative
standing of
the
measurement
s. Table 5.5
shows
the
standardized
canonical
discriminant
function
coefficients.
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
TABLE 5.5 STANDARDIZED CANONICAL DISCRIMINANT FUNCTION TABLE 5.6 CANONICAL DISCRIMINANT FUNCTION COEFFICIENTS COEFFICIENTS
Interpretation
From the table 5.5 it can be concluded
that the financial factors such as Operating
Profit, Quick Ratio, Interest Coverage
Ratio, Working Capital Turnover Ratio,
Inventory Turnover Ratio, and Net Working
Capital shall be considered significant in
classifying the applicants.
Interpretation
The territorial map is drawn with the
first discriminant function representing the x
axis and the second discriminant function
representing the y axis.
The discriminant score of each case is
plotted in the territorial map. Also the
discriminant scores at the centroid of each
group are plotted. A case is predicted as
Interpretation
These coefficients give
the exact weightage of
each variable to arrive at
discriminant scores.
E. Discriminant
Functions
1 A
discriminant
function, also called
a canonical root, is a
latent variable which
is created as a linear
combination
of
discriminating
(independent)
variables.
2 There
is
one
discriminant function
for
2-group
discriminant
analysis, but for
higher
order
discriminant
Analysis, the number
of functions is (g 1), where g is the
number of categories
in
the
grouping
variable.
3 The first function
will be the most
powerful
differentiating
dimension, but later
functions may also
represent additional
significant
dimensions
of
differentiation.
Discriminant Function 1
Y = - 0.031 *
Existence - 0.657 *
Location + 0.124 * Group 0.112
*
Distribution
Network - 0.227 *
Relationship
with
Customers - 0.175 *
371
1*
2*
PAT/Total Asset
+ 0.045 * PAT/Net
worth - 0.010 *
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
F. Discriminant Score
classification for this
The discriminant score, alsosample. Overall 88.0 %
called the Discriminant Analysis
of selected cases (i.e.,
score, is the value resulting from
the cases used for
applying a discriminant function
building the model)
formula to the data for a given
were
correctly
case.
classified.
Classification Function
G. Validation of the
Coefficients
The classification function Model
coefficients can also be used to For checking the
predict the group membership.
consistency
of
TABLE 5.7 CLASSIFICATION
performance of the
FUNCTION COEFFICIENTS
model, twenty four
existing clients were
considered and rated.
The comparison of the
rating of the customer
and the current status
of the performance
determines
the
consistency of the
model. Table 5.9 shows
the validation results.
TABLE
5.9
VALIDA
TION
RESULTS
Interpretation
column
contains
estimates of the coefficients
for a classification function
for one group.
2 The functions are used to
assign or classify cases into
groups.
3 To obtain a classification
score for each case for each
group,
multiply
each
coefficient by the value of the
corresponding variable, sum
the products, and add the
constant to get the score.
4 A case is predicted as being a
member of the group in
which the value of its
classification function is
largest.
Classification of Applicants
1 Each
Interpretation
20 cases which were
not included in the
model building were
used for the validation
for
model.
The
accuracy
of
classification is 90%.
Hence, the model is
highly reliable.
VI. FINDINGS
AND
SUGGESTIONSKEY
FINDINGS
28 factors (Financial
and
Non-financial)
were considered for
building the model and
the
classification
accuracy produced by
the model was 90%.
The factors cannot
be given equal weights.
The level of impact of
each
factor
in
predicting the risk
category
of
an
applicant differs.
Among the financial
factors considered for
building the model,
Operating
Profit,
Quick Ratio, Interest
Coverage
Ratio,
Working
Capital
Turnover
Ratio,
Inventory
Turnover
Ratio,
and
Net
Working Capital shall
be
considered
significant
in
classifying
the
applicants.
Apart from financial
factors, non-financial
factors such as Track
Record
of
Management,
The
Group
of
the
Company,
Product
Range, Access to
Human
Resources,
and Relationship with
International Journal of Innovation, Management and Technology, Vol. 1, No. 4, October 2010
ISSN: 2010-0248
[1]
[2]
[3]
373