Simple example of Credit Risk Model-as-a-service API using Flask-Restplus
The sample model given here is the KMV model. The KMV model calculates the firm's expected default frequency (EDF) of a firm. EDF is also a proxy for probability-of-default. Credit to(HK UST) for the actual presentation - https://www.math.ust.hk/~maykwok/Web_ppt/KMV/KMV.pdf
def kmv(enterprise_value, short_term_debt, long_term_debt, mu, sigma, period=1):
"""
KMV Model - https://www.math.ust.hk/~maykwok/Web_ppt/KMV/KMV.pdf
:param enterprise_value: Enterprise Value of the Firm (can market capitalisation)
:param short_term_debt: Firm's short term debt
:param long_term_debt: Firm's long term debt
:param mu: Expected Return after 1 year
:param sigma: Expected Annualized Volatility
:param period: period in years
:return: EDF (Expected Default Frequency or Probability of Default)
"""
# Calculate default point
default_point = short_term_debt + (0.5 * long_term_debt)
# Numerator
numer = math.log(enterprise_value / default_point) + (mu - math.pow(sigma, 2) / 2) * period
# Denominator
denom = sigma * period
# Distance to Default
distance_to_default = numer / denom
edf = norm.cdf(-distance_to_default)
return edf
