Merton-model Approach to Valuing Correlation Products

Similar documents
Chapter 5 Student Lecture Notes 5-1

Structural Models I. Viral V. Acharya and Stephen M Schaefer NYU-Stern and London Business School (LBS), and LBS. Credit Risk Elective Spring 2009

Final Exam. 7. (10 points) Please state whether each of the following statements is true or false. No explanation needed.

DOUBLE IMPACT. Credit Risk Assessment for Secured Loans. Jean-Paul Laurent ISFA Actuarial School University of Lyon & BNP Paribas

Elton, Gruber, Brown and Goetzmann. Modern Portfolio Theory and Investment Analysis, 7th Edition. Solutions to Text Problems: Chapter 4

Empirical estimation of default and asset correlation of large corporates and banks in India

4. Greek Letters, Value-at-Risk

Correlations and Copulas

Data Mining Linear and Logistic Regression

Problem Set 6 Finance 1,

Chapter 15: Debt and Taxes

Comparative analysis of CDO pricing models

Linear Combinations of Random Variables and Sampling (100 points)

Risk and Return: The Security Markets Line

Maturity Effect on Risk Measure in a Ratings-Based Default-Mode Model

Chapter 6 Risk, Return, and the Capital Asset Pricing Model

Multifactor Term Structure Models

Basket options and implied correlations: a closed form approach

Clearing Notice SIX x-clear Ltd

Asian basket options. in oil markets

Impact of CDO Tranches on Economic Capital of Credit Portfolios

Elton, Gruber, Brown, and Goetzmann. Modern Portfolio Theory and Investment Analysis, 7th Edition. Solutions to Text Problems: Chapter 9

How Likely Is Contagion in Financial Networks?

II. Random Variables. Variable Types. Variables Map Outcomes to Numbers

Midterm Exam. Use the end of month price data for the S&P 500 index in the table below to answer the following questions.

Basel II Credit Loss Distributions under Non-Normality

A Set of new Stochastic Trend Models

Chapter 3 Descriptive Statistics: Numerical Measures Part B

Economics 1410 Fall Section 7 Notes 1. Define the tax in a flexible way using T (z), where z is the income reported by the agent.

Physics 4A. Error Analysis or Experimental Uncertainty. Error

Network Analytics in Finance

Networks in Finance and Marketing I

Notes on experimental uncertainties and their propagation

Topic 6 Introduction to Portfolio Theory

FORD MOTOR CREDIT COMPANY SUGGESTED ANSWERS. Richard M. Levich. New York University Stern School of Business. Revised, February 1999

The Hiring Problem. Informationsteknologi. Institutionen för informationsteknologi

Modelling the Credit Risk for Portfolios of Consumer Loans: Analogies with corporate loan models

Prospect Theory and Asset Prices

Option pricing and numéraires

Tests for Two Correlations

Survey of Math Test #3 Practice Questions Page 1 of 5

MgtOp 215 Chapter 13 Dr. Ahn

CHAPTER 9 FUNCTIONAL FORMS OF REGRESSION MODELS

In the 1990s, Japanese economy has experienced a surge in the unemployment rate,

THE VOLATILITY OF EQUITY MUTUAL FUND RETURNS

Appendix - Normally Distributed Admissible Choices are Optimal

Lecture 6 Foundations of Finance. Lecture 6: The Intertemporal CAPM (ICAPM): A Multifactor Model and Empirical Evidence

Examining the Validity of Credit Ratings Assigned to Credit Derivatives

2) In the medium-run/long-run, a decrease in the budget deficit will produce:

Lecture 10: Valuation Models (with an Introduction to Capital Budgeting).

Tests for Two Ordered Categorical Variables

To Rebalance or Not to Rebalance? Edward Qian, PhD, CFA PanAgora Asset Management

Teaching Note on Factor Model with a View --- A tutorial. This version: May 15, Prepared by Zhi Da *

/ Computational Genomics. Normalization

Financial Stability and Basel II

CDO modelling from a practitioner s point of view: What are the real problems? Jens Lund 7 March 2007

Notes are not permitted in this examination. Do not turn over until you are told to do so by the Invigilator.

3/3/2014. CDS M Phil Econometrics. Vijayamohanan Pillai N. Truncated standard normal distribution for a = 0.5, 0, and 0.5. CDS Mphil Econometrics

Model Study about the Applicability of the Chain Ladder Method. Magda Schiegl. ASTIN 2011, Madrid

THE RELATIONSHIP BETWEEN AVERAGE ASSET CORRELATION AND DEFAULT PROBABILITY

Labor Market Transitions in Peru

Introduction. Chapter 7 - An Introduction to Portfolio Management

Digital assets are investments with

Economic Design of Short-Run CSP-1 Plan Under Linear Inspection Cost

A Credit Risk Model Incorporating Microstructural Dependencies and Stochastic Recovery

Jean-Paul Murara, Västeras, 26-April Mälardalen University, Sweden. Pricing EO under 2-dim. B S PDE by. using the Crank-Nicolson Method

arxiv: v1 [q-fin.pm] 13 Feb 2018

RISK-BASED CAPITAL AND CREDIT INSURANCE PORTFOLIOS *

arxiv:cond-mat/ v1 [cond-mat.other] 28 Nov 2004

REFINITIV INDICES PRIVATE EQUITY BUYOUT INDEX METHODOLOGY

Principles of Finance

Sharing Risk An Economic Perspective 36th ASTIN Colloquium, Zurich, Andreas Kull, Global Financial Services Risk Management

TCOM501 Networking: Theory & Fundamentals Final Examination Professor Yannis A. Korilis April 26, 2002

Introduction to PGMs: Discrete Variables. Sargur Srihari

Consumption Based Asset Pricing

Random Variables. b 2.

Credit Name Concentration Risk: Granularity Adjustment Approximation

Random Variables. 8.1 What is a Random Variable? Announcements: Chapter 8

Test Bank to accompany Modern Portfolio Theory and Investment Analysis, 9 th Edition

General Examination in Microeconomic Theory. Fall You have FOUR hours. 2. Answer all questions

Chapter 10 Making Choices: The Method, MARR, and Multiple Attributes

σ may be counterbalanced by a larger

Chapter 5 Risk and return

Elements of Economic Analysis II Lecture VI: Industry Supply

How diversifiable is firm-specific risk? James Bennett. and. Richard W. Sias * October 20, 2006

The sustainability of credit to low income mortgagors: a risk management perspective

3: Central Limit Theorem, Systematic Errors

Risk Quantification of Retail Credit: Current Practices and Future Challenges*

A Theory of Predation Based on Agency Problems in Financial Contracting

Analysis of Variance and Design of Experiments-II

Risk Integrated

Mode is the value which occurs most frequency. The mode may not exist, and even if it does, it may not be unique.

Mutual Funds and Management Styles. Active Portfolio Management

Measures of Spread IQR and Deviation. For exam X, calculate the mean, median and mode. For exam Y, calculate the mean, median and mode.

25.1. Arbitrage Pricing Theory Introduction

THE ISSUE OF PD ESTIMATION A PRACTICAL APPROACH

Global sensitivity analysis of credit risk portfolios

Chapter 5 Bonds, Bond Prices and the Determination of Interest Rates

Mixed-Integer Credit Portfolio Optimization: an application to Italian segregated funds

Investment Management Active Portfolio Management

Transcription:

Merton-model Approach to Valung Correlaton Products Vral Acharya & Stephen M Schaefer NYU-Stern and London Busness School, London Busness School Credt Rsk Electve Sprng 2009 Acharya & Schaefer: Merton model approach for correlaton products 1

Bnomal wth Merton Model Important method for calculatng dstrbuton of loan losses: wdely used n bankng used by Basel II regulatons to set bank captal requrements Lnked to dstance-to-default analyss Acharya & Schaefer: Merton model approach for correlaton products 2

Mxed Bnomal: Usng Merton s Models as Mxng Dstrbuton In Merton model value of rsky debt depends on frm value and default rsk s correlated because frm values are correlated (e.g., va common dependence on market factor). Value of frm at tme T: V = V exp( ( µ (1/ 2) 2 ) ) where ~ (0,1), σ T + σ T % ε %, V, ε N T V 1 4 44 2 4 4 43 14 2 43 surprse n R C expected value of R C We wll assume that correlaton between frm values arses because of correlaton between surprse n ndvdual frm value (ε ι ) and market factor (m) Acharya & Schaefer: Merton model approach for correlaton products 3

Mxed Bnomal: Usng Merton s Models as Mxng Dstrbuton Suppose correlaton between each frm s value and the market factor s the same and equal to sqrt( ). Ths means that we may model correlaton between the ε s as ε = m + 1 v, = 1, K N and corr( ε, ε ) = j Where m and v are ndependent N(0,1) random varables and s common to all frms Notce that f v ~ N(0,1) and m ~ N(0,1) then ε ~ N(0,1) Acharya & Schaefer: Merton model approach for correlaton products 4

Structural Approach, contd. From our analyss of dstance-to-default, we know that under the Merton Model a frm defaults when: 1 2 ε R ( µ σ ) T / σ T where = R ln( B / V ),,,, 2 V D V D The uncondtonal probablty of default, p, s therefore: p < Prob ε R ( µ σ ) ( µ σ ) = N σ T σ T 1 2 1 2 D, 2 V, RD 2 V, V, In ths model we assume that the default probablty, p, s constant across frms Acharya & Schaefer: Merton model approach for correlaton products 5

Structural Approach to Correlaton the Idea Workng out the dstrbuton of portfolo losses drectly when the ε s are correlated s not easy But, f we work out the dstrbuton condtonal on the market shock, m, then we can explot the fact that the remanng shocks are ndependent and work out the portfolo loss dstrbuton Acharya & Schaefer: Merton model approach for correlaton products 6

Structural Approach, contd. The shock to the return, ε, s related to the common and dosyncratc shocks by: Default occurs when: 1 2 RD, ( µ 2 σv, ) ε = m + 1 v < = N ( p) σ T or v < N ( p) m 1 ε = m + 1 v V, Acharya & Schaefer: Merton model approach for correlaton products 7

The Default Condton v < N 1 ( p) m 1 A large value of m means a good shock to the market (hgh asset values) The larger the value of m the more negatve the dosyncratc shock, v, has to be to trgger default The hgher the correlaton,, between the frm shocks, the larger the mpact of m on the crtcal value of v. Acharya & Schaefer: Merton model approach for correlaton products 8

Structural Approach, contd. Condtonal on the realsaton of the common shock, m, the probablty of default s therefore: Prob(default m)= Prob v < N ( p) m 1 N ( p) m = N = θ ( m), say 1 N ( p) m N = and therfore = ( θ ) 1 Acharya & Schaefer: Merton model approach for correlaton products 9

The relaton between m and θ For a gven market shock, m, θ gves the condtonal probablty of default on an ndvdual loan 0.25 Hgh Corr p = 6.0% rho = 25.0% Theta (condtonal default prob) 0.20 0.15 0.10 0.05 Low Corr p = 6.0% rho = 5.0% 0.00-4 -3-2 -1 0 1 2 3 4 market shock (m) Acharya & Schaefer: Merton model approach for correlaton products 10

Implcatons of Condtonal Independence For a gven value of m, as the number of loans n the portfolo, the proporton of loans n the portfolo that default converges to the probablty θ The probablty that the loan-loss proporton, L, s < θ s therefore: N ( p) m Prob( L = θ ) Prob N ( θ ) 1 1 = Prob m N ( p) N ( ) 1 ( ) θ 1 = Prob m 1 N ( ) N ( p) ( ) θ 1 = N 1 ( ) ( ) ( ) N θ N p Acharya & Schaefer: Merton model approach for correlaton products 11

Loan Loss Dstrbuton Structural Model Ths result gves the dstrbuton of the fracton of loans that default n a well dversfed homogeneous portfolo where the correlaton n default comes from dependence on a common factor Homogenety means that each loan has: the same default probablty, p (mplctly) the same loss-gven-default the same correlaton,, across dfferent loans The dstrbuton has two parameters 1 Prob( L = θ ) N 1 N ( θ ) N ( p) default probablty, p correlaton, ( ) Acharya & Schaefer: Merton model approach for correlaton products 12

Loan Loss Dstrbuton wth p = 1% and = 12% and 0.6% 0.08 p = 1.5% rho = 12.0% 0.07 p = 1.5% rho = 0.6% 0.06 0.05 0.04 0.03 0.02 0.01 0 0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% Portfolo Loan Loss (%( Acharya & Schaefer: Merton model approach for correlaton products 13

Example of Vascek formula Appled to Bank Portfolo Source: Vascek Acharya & Schaefer: Merton model approach for correlaton products 14