Professional Documents
Culture Documents
Models
Bappaditya Mukhopadhyay
Management Development Institute,
2002
Understanding and Modeling
Default Risks
Linear Probability Models
Logistic Models
Other Models
Traditional Approach:
Credit Scoring System: Altman’s Z – Score
model
Z= 1.2 X1+ 1.4 X2 + 3.3 X3 + 0.6 X4+ 1.0 X5
X1: Working Capital/ Total assets
X2: Retained Earnings/Total assets
X3: EBIT/Total assets
X4: Market Value of Equity/ Total Assets
X5: Sales/ Total Assets
Issues
`Cut off’ value of Z: 1.81
Works well under normal conditions
Suffers from standard Econometrics
Problems?…. Multicollinearity….
….Therefore prone to type I and type II
errors
Loans as Options: The KMV Model
What is EDF?
Suppose Mean Asset Value, A = 100 crores
St. Deviation Asset Value,σA = 10 crores
Value of Debt (Face value), B = 80 crores
Advantages:
Reflects information signals
Applicable to any public company
Structural Model
Disadvantages:
Assumption of normality
What about non listed ones
Constant debt structure
Problems with BS model...
Merton’s Valuation Model
d: Leverage ratio
h1: - [ ½ 2(T-t) – ln (d) ]/ (T-t)
h2: - [ ½ 2(T-t) + ln (d) ]/ (T-t)
A Numerical Example:
B = 100,000
(T-t)= 1 year
i = 5%
d = 0.9
σ = 12%
Rating Migration
Stable Markov Process: `` No serial correlation with
previous movements”
Transition matrix same across borrower types
Bond aging?
Bond Covenants?
McKinsey Model: Macro Simulation
Three Approaches:
Simulations Approach
New Challenges and Way Ahead