Advanced Tools for Risk Management and Asset Pricing

Similar documents
Models of Default Risk

INSTITUTE OF ACTUARIES OF INDIA

Brownian motion. Since σ is not random, we can conclude from Example sheet 3, Problem 1, that

Matematisk statistik Tentamen: kl FMS170/MASM19 Prissättning av Derivattillgångar, 9 hp Lunds tekniska högskola. Solution.

Alexander L. Baranovski, Carsten von Lieres and André Wilch 18. May 2009/Eurobanking 2009

Jarrow-Lando-Turnbull model

Synthetic CDO s and Basket Default Swaps in a Fixed Income Credit Portfolio

INSTITUTE OF ACTUARIES OF INDIA

UCLA Department of Economics Fall PhD. Qualifying Exam in Macroeconomic Theory

Financial Econometrics Jeffrey R. Russell Midterm Winter 2011

Hull-White one factor model Version

The Mathematics Of Stock Option Valuation - Part Four Deriving The Black-Scholes Model Via Partial Differential Equations

MA Advanced Macro, 2016 (Karl Whelan) 1

Market Models. Practitioner Course: Interest Rate Models. John Dodson. March 29, 2009

Introduction to Black-Scholes Model

MAFS Quantitative Modeling of Derivative Securities

FINAL EXAM EC26102: MONEY, BANKING AND FINANCIAL MARKETS MAY 11, 2004

(1 + Nominal Yield) = (1 + Real Yield) (1 + Expected Inflation Rate) (1 + Inflation Risk Premium)

Research Article A General Gaussian Interest Rate Model Consistent with the Current Term Structure

VaR and Low Interest Rates

Pricing FX Target Redemption Forward under. Regime Switching Model

Agenda. What is an ESG? GIRO Convention September 2008 Hilton Sorrento Palace

May 2007 Exam MFE Solutions 1. Answer = (B)

Fundamental Basic. Fundamentals. Fundamental PV Principle. Time Value of Money. Fundamental. Chapter 2. How to Calculate Present Values

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

Proceedings of the 48th European Study Group Mathematics with Industry 1

CENTRO DE ESTUDIOS MONETARIOS Y FINANCIEROS T. J. KEHOE MACROECONOMICS I WINTER 2011 PROBLEM SET #6

Principles of Finance CONTENTS

Option Valuation of Oil & Gas E&P Projects by Futures Term Structure Approach. Hidetaka (Hugh) Nakaoka

Description of the CBOE S&P 500 2% OTM BuyWrite Index (BXY SM )

Constructing Out-of-the-Money Longevity Hedges Using Parametric Mortality Indexes. Johnny Li

INSTITUTE OF ACTUARIES OF INDIA

Tentamen i 5B1575 Finansiella Derivat. Måndag 27 augusti 2007 kl Answers and suggestions for solutions.

A pricing model for the Guaranteed Lifelong Withdrawal Benefit Option

Description of the CBOE Russell 2000 BuyWrite Index (BXR SM )

Appendix B: DETAILS ABOUT THE SIMULATION MODEL. contained in lookup tables that are all calculated on an auxiliary spreadsheet.

Optimal Early Exercise of Vulnerable American Options

IJRSS Volume 2, Issue 2 ISSN:

MORNING SESSION. Date: Wednesday, October 30, 2013 Time: 8:30 a.m. 11:45 a.m. INSTRUCTIONS TO CANDIDATES

MORNING SESSION. Date: Wednesday, April 26, 2017 Time: 8:30 a.m. 11:45 a.m. INSTRUCTIONS TO CANDIDATES

7 pages 1. Hull and White Generalized model. Ismail Laachir. March 1, Model Presentation 1

Tentamen i 5B1575 Finansiella Derivat. Torsdag 25 augusti 2005 kl

Comparison of back-testing results for various VaR estimation methods. Aleš Kresta, ICSP 2013, Bergamo 8 th July, 2013

Pricing Vulnerable American Options. April 16, Peter Klein. and. Jun (James) Yang. Simon Fraser University. Burnaby, B.C. V5A 1S6.

Systemic Risk Illustrated

Ch. 10 Measuring FX Exposure. Is Exchange Rate Risk Relevant? MNCs Take on FX Risk

Origins of currency swaps

EVA NOPAT Capital charges ( = WACC * Invested Capital) = EVA [1 P] each

Foreign Exchange, ADR s and Quanto-Securities

HEDGING VOLATILITY RISK

You should turn in (at least) FOUR bluebooks, one (or more, if needed) bluebook(s) for each question.

Black-Scholes Model and Risk Neutral Pricing

DEBT INSTRUMENTS AND MARKETS

Financial Econometrics (FinMetrics02) Returns, Yields, Compounding, and Horizon

Data Mining Anomaly Detection. Lecture Notes for Chapter 10. Introduction to Data Mining

Data Mining Anomaly Detection. Lecture Notes for Chapter 10. Introduction to Data Mining

Problem Set 1 Answers. a. The computer is a final good produced and sold in Hence, 2006 GDP increases by $2,000.

Quanto Options. Uwe Wystup. MathFinance AG Waldems, Germany 19 September 2008

Option pricing and hedging in jump diffusion models

Computations in the Hull-White Model

Economic Growth Continued: From Solow to Ramsey

A UNIFIED PDE MODELLING FOR CVA AND FVA

ANSWER ALL QUESTIONS. CHAPTERS 6-9; (Blanchard)

Single Premium of Equity-Linked with CRR and CIR Binomial Tree

STATIONERY REQUIREMENTS SPECIAL REQUIREMENTS 20 Page booklet List of statistical formulae New Cambridge Elementary Statistical Tables

Available online at ScienceDirect

Online Appendix. Using the reduced-form model notation proposed by Doshi, el al. (2013), 1. and Et

Advanced Tools for Risk Management and Asset Pricing

Financial Markets And Empirical Regularities An Introduction to Financial Econometrics

A Note on Missing Data Effects on the Hausman (1978) Simultaneity Test:

Pricing formula for power quanto options with each type of payoffs at maturity

4452 Mathematical Modeling Lecture 17: Modeling of Data: Linear Regression

ECONOMIC GROWTH. Student Assessment. Macroeconomics II. Class 1

Supplement to Chapter 3

Macroeconomics II A dynamic approach to short run economic fluctuations. The DAD/DAS model.

Leveraged Stock Portfolios over Long Holding Periods: A Continuous Time Model. Dale L. Domian, Marie D. Racine, and Craig A.

Mathematical methods for finance (preparatory course) Simple numerical examples on bond basics

DYNAMIC ECONOMETRIC MODELS Vol. 7 Nicolaus Copernicus University Toruń Krzysztof Jajuga Wrocław University of Economics

Technological progress breakthrough inventions. Dr hab. Joanna Siwińska-Gorzelak

Lecture Notes to Finansiella Derivat (5B1575) VT Note 1: No Arbitrage Pricing

San Francisco State University ECON 560 Summer 2018 Problem set 3 Due Monday, July 23

On the Impact of Inflation and Exchange Rate on Conditional Stock Market Volatility: A Re-Assessment

Risk-Neutral Probabilities Explained

1 Purpose of the paper

CURRENCY CHOICES IN VALUATION AND THE INTEREST PARITY AND PURCHASING POWER PARITY THEORIES DR. GUILLERMO L. DUMRAUF

1. Interest Rate Gap. Duration

Optimal Consumption and Investment with Habit Formation and Hyperbolic discounting. Mihail Zervos Department of Mathematics London School of Economics

Standard derivatives pricing theory (see, for example, Hull,

Change of measure and Girsanov theorem

Funding beyond discounting: collateral agreements and derivatives pricing

SOCIETY OF ACTUARIES Quantitative Finance and Investment Advanced Exam Exam QFIADV MORNING SESSION

Subdivided Research on the Inflation-hedging Ability of Residential Property: A Case of Hong Kong

FIXED INCOME MICHAEL MONOYIOS

Erratic Price, Smooth Dividend. Variance Bounds. Present Value. Ex Post Rational Price. Standard and Poor s Composite Stock-Price Index

Interest rate models enhanced with local volatility

HEDGING VOLATILITY RISK

An Analytical Implementation of the Hull and White Model

CHRISTOPH MÖHR ABSTRACT

A Method for Estimating the Change in Terminal Value Required to Increase IRR

Affine Term Structure Pricing with Bond Supply As Factors

Transcription:

MSc. Finance/CLEFIN 214/215 Ediion Advanced Tools for Risk Managemen and Asse Pricing May 215 Exam for Non-Aending Sudens Soluions Time Allowed: 13 minues Family Name (Surname) Firs Name Suden Number (Mar.) Please answer all quesions by choosing he mos appropriae alernaive and/or by wriing your answers in he spaces provided. You need o carefully jusify and show your work in he case of open quesions. There is only one correc answer(s) for each of he muliple choice quesions: each seleced alernaive ha is correc will be awarded one poin. Only answers explicily repored in he appropriae box will be considered. No oher answers or indicaions poining o poenial answers will be aken ino consideraion. In he case of open quesions, he maximum number of poins is indicaed. Quesion 1. Which of he following saemens abou Correlaion is FALSE? (A) Correlaion is well defined for heavy-ailed disribuions (B) Correlaion measures he linear dependence and akes values in he inerval [-1,1] (C) Correlaion is invarian under sricly increasing linear ransformaions (D) Correlaion is linked boh o he copula funcion and o he marginal disribuions Quesion 2. Which of he following funcions can NOT be a copula funcion? (A) C(u₁, u₂) = u₁u₂ (B) C(u₁, u₂) = min(u₁, u₂) (C) C(u₁, u₂) = max(u₁ + u₂, ) (D) None of he above 1

Quesion 3. Which of he following saemens abou Rank Correlaions is FALSE? (A) To compue rank correlaions one needs o know boh he numerical values of he variables and he ordering of he sample for each variable (B) Rank correlaions are invarian under sricly increasing ransformaions (C) Rank correlaions ake a value of -1 when he variables are counermonoonic and he value 1 when he variables are comonoonic (D) The populaion version of Kendall's au can be simply expressed as: ρ τ (X₁, X₂) = E(sign((X₁ X ₁)(X₂ X ₂))) Quesion 4. Which of he following saemens abou Copula Esimaion and Calibraion is FALSE? (A) A Gaussian copula is asympoically independen in boh ails (B) A Mehod-of-Momens approach involves calibraion of a copula using empirical esimaes for some rank correlaion measures (C) Boh Gumbel and Clayon copulas have upper ail dependence (D) Measures of exremal dependence beween a pair of random variables ₁ and X X₂ depend only on he copula of X₁ and X₂ Quesion 5. Which of he following saemens abou Sochasic Processes is TRUE? (A) Io's Inegral is a maringale (B) A Brownian moion is a saionary process wih independen Gaussian incremens. (C) The sandard chain rule does NOT work for Io's Inegral. (D) All he above saemens are TRUE Quesion 6. In Vasicek's ineres rae model, he dynamics of he shor rae process (r ) are given by he SDE dr = k(θ r )d + σdw, where k, θ and σ are sricly posiive consans. Show ha he soluion of SDE above is given by r = θ + (r θ)e k + σe k e ks dw s Hin: Consider he Io processes (X ) and (Y ) defined by X = e k and Y = r and use inegraion by pars. Answer. a) We consider he Io process (X ) and (Y ) defined by X = e k and Y = r and we noe ha 2

dx = ke k d, X₀ = 1, dy = k(θ r )d + σdw, Y₀ = r₀ Using he inegraion by pars formula, we calculae This implies ha e k r X Y = X₀Y₀ + Y s dx s + X s dy s = X₀Y₀ + [ke ks Y s + k(θ r )X s ]ds + (dy s )(dx s ) + σx s dw s = r₀ + [ke ks r s + k(θ r s )e ks ]ds + σe ks dw s = r₀ + θe k θ + σ e ks dw s. r = θ + (r θ)e k + σe k σe ks dw s. (b) Noing ha he sochasic inegral wih respec o he Brownian moion has expecaion, we can see ha he expecaion of he random variable r_{} is given by Using Io's isomery, we calculae E[r ] = θ + (r θ)e k var(r ) = E[(r E[r ])²] = σ 2 e 2k E e ks dw s = ((σ²)/(2k))(1 e 2k ) 2 Quesion 7. Assume a CDS quoed spread is 3 basis poins and he recovery rae is esimaed o be 4%. Under he assumpions ha i) he premium leg of he CDS pays coninuously and ii) he hazard rae is consan, wha is he value of he hazard rae? (A) 5% (B) 7.5% (C) 5 (D) 75 3

Quesion 8. The Gaussian Copula Approach allows o compue he join probabiliy of defaul of n names. Such probabiliy enails i) he calculaion of a muli-dimensional inegral and ii) he esimaion of he correlaion marix among names. Given ha dim I = dimension of he inegral and dim ρ = number of free correlaion parameers (enries of he correlaion marix), wha are he values of hese wo parameers under he Single-Facor Gaussian Copula Approach? (A) dim I = n, dim ρ = n(n 1) 2 (B) dim I = 1, dim ρ = n(n 1) 2 (C) dim I = n, dim ρ = n (D) dim I =1, dim ρ = n Quesion 9. Which of he following saemens abou implied correlaions is rue? (A) The compound correlaion is inconsisen a he level of a single ranche (B) The compound correlaion for a given ranche is always unique (C) Base correlaion can yield negaive expeced ranche losses (D)Base correlaion depends on pairs of aachmen poins Quesion 1. Which of he following saemens abou credi modeling in a muli-facor se up is false? (A) Secor concenraion risk is due o he correlaion srucure of obligors inside he porfolio (B) Secor concenraion risk affecs only he condiional variance erm (C) Name concenraion risk is a second order effec (D) Name concenraion risk can be diversified away Quesion 11. Which of he following saemens abou he Limiing Porfolio Loss Disribuion, inroduced by Vasicek, is false? (A) By he law of large numbers, for an infiniely granular porfolio, he porfolio loss disribuion converges o is expecaion (B) The limiing loss disribuion provides a good approximaion o he porfolio loss, also for a porfolio consising of uneven (bu no dominaing) weighs (C) The probabiliy densiy funcion of he porfolio loss can be U-shaped (D) The probabiliy densiy funcion of he porfolio loss can be unimodal 4

Quesion 12. Which of he following saemens abou differen approaches o price CDOs is rue? (A) The Implied Copula Approach by Hull and Whie solves he problems of inconsisencies across he capial srucure and mauriies (B) The Gaussian Copula Approach is able o capure he phenomenon of clusered (secor) defauls associaed o masses in he far righ ail of he loss disribuion (C) Dynamic(al) Loss Approaches model direcly aggregaed objecs (D) Dynamic(al) Loss Approaches allow o calculae single name sensiiviies Quesion 13. Which of he following saemens abou Analyically Tracable Firs Passage (AT1P) models is false? (A) AT1P models assume he exisence of a ime-dependen deerminisic barrier (B) AT1P models poses he problem of he correc deerminaion of shor erm credi spreads (C) AT1P models implicily assume ha i is possible o reach a considerable probabiliy of defaul in a very shor ime horizon smoohly (D)AT1P models implicily assume ha accouning daa are fully reliable and ransparen Quesion 14. Which of he following saemens abou Reduced Form (Inensiy) models is rue? (A) Sochasic inensiy models always allow o obain large levels of opion implied volailiies for CDS raes (B) In a muli-name seing, inroducing dependence in sochasic inensiies of differen names can lead o unrealisically low levels of dependence across defauls (C) Negaive values of he inensiy can never be aained in CDS calibraion (D) Calibraion o CDS spreads is more sable wih piece-wise linear inensiies compared o piece-wise consan ones Quesion 15. Which of he following saemens abou Reduced Form (Inensiy) models is false? (A) Cox processes allow o ake ino consideraion spread volailiy (B) Defaul is described by an endogenous jump process (C) In deerminisic inensiy models, survival probabiliies have he same srucure as discoun facors in shor rae models (D) Survival probabiliies obained hrough calibraion o CDS quoes can increase wih ime 5

Quesion 16. Consider a bespoke CDO ha conains he same credis as he index porfolio. The recovery rae for he index is 6% and for he bespoke i is. Which of he following saemens is false? (A) A 1% ranche on he bespoke will experience he same relaive losses as a 6% ranche on he index (B) A 1% ranche on he bespoke will experience he same relaive losses as a 4% ranche on he index (C) Losses on he bespoke are 2.5 imes he losses on he index (D) The 1% srike on he bespoke should be priced wih he same correlaion as he 4% srike on he index Quesion 17. Wihin he framework of srucural models of defaul, he firs model had been inroduced by Meron in 1974. Briefly describe he assumpions underlying he model, is limis and exensions (wrie a mos 2 lines of ex). Answer. See Lecure 9 Srucural Models : Assumpions: slide 8/65 Limis: slide 21/65 Exensions: slides 22-23/65 Quesion 18. In he framework of he Gaussian Copula approach, wo ypes of implied correlaion, i.e., compound correlaion and base correlaion have been inroduced. Briefly describe hem and heir limiaions (wrie a mos 2 lines of ex). Answer. See Lecure 1 CDO Basics Compound correlaion I is based on he assumpion ha each ranche [A,B] is characerized by a unique value of correlaion ρ AB. Typically, he compound correlaion srucure presens a smile. Limiaions: 1. I canno be easily inerpolaed/exrapolaed 2. I is unable o price bespoke ranches 3. I may no exis Base correlaion I is based on he assumpion ha each equiy ranche [,X] is characerized by a unique value of correlaion ρ X. I follows ha a ranche [A,B] depends on wo values of base correlaion. Typically, he compound correlaion srucure presens a skew. Limiaions: 1. I is inconsisen a he level of single ranche 2. I may yield negaive expeced ranche losses 6

Quesion 19. Considering he wo ermshees below, which of he following saemens is plausible: Bonus Cap A Bonus Cap B Underlying Fia Underlying Fia Mauriy 3 Years Mauriy 3 Years Barrier 7% Barrier 7% Barrier ype American Barrier ype European (A) Bonus Cap A has a Bonus equal o 115% and Bonus Cap B has a Bonus equal o 12%. (B) Bonus Cap A has a Bonus equal o 11% and Bonus Cap B has a Bonus equal o 18%. (C) Bonus Cap A has a Bonus equal o 18% and Bonus Cap B has a Bonus equal o 18%. (D) Bonus Cap A has a Bonus equal o 98% and Bonus Cap B has a Bonus equal o 95%. Quesion 2. A Shor Equiy Proecion which pays Eur 1 if a mauriy he underlying (wih price S()) is above he Srike and Eur 1 [1 + (1 S(T)/S() ) ] if he underlying is below he Srike is replicaed by: (A) Buying a Zero Coupon Bond and selling an ATM pu opion (B) Buying a Zero Coupon Bond and buying an ATM call opion (C) Buying a Zero Coupon Bond and selling an ATM call opion (D) Buying a Zero Coupon Bond and buying an ATM pu opion Quesion 21. You are srucuring an Equiy Proecion cerificae wih 1% capial proecion; he Zero Coupon Bond coss Eur 95. An ATM call opion on he FTSE MIB index coss 7 Euro. Because you would like your produc o offer 1% paricipaion o any poenial appreciaion of he underlying, which sraegy of selecion of an underlying differen from he FTSE MIB index would you consider? (A) I shall no need any alernaive selecion of he underlying asse because I can already offer 1% paricipaion o any poenial appreciaion of he underlying (B) I will be looking for an underlying asse wih lower volailiy and a higher dividend yield han he FTSE MIB so ha he opion will be cheaper o ry and aim a an opion cos of Eur 5 o make a 1% proecion possible (C) I will be looking for an underlying asse wih lower volailiy and dividend yield han he FTSE MIB so ha he opion will be cheaper o ry and aim a an opion cos of Eur 5 o make a 1% proecion possible (D) I will be looking for an underlying asse wih higher volailiy and dividend yield han he FTSE MIB so ha he opion will be cheaper o ry and aim a an opion cos of Eur 5 o make a 1% proecion possible 7

Quesion 22. An ouperformance cerificae is: (A) A cerificae wih capial proecion and one-o-one paricipaion o poenial appreciaion of he underlying (B) A cerificae wih capial proecion and magnified paricipaion o poenial appreciaion of he underlying (C) A cerificae wih condiional capial proecion and one-o-one paricipaion o poenial appreciaion of he underlying (D) None of he above Quesion 23. Which of he opions lised below will be he cheapes: (A) An ATM Asian opion on he FTSE MIB wih 3-year mauriy (B) An ATM European opion on he FTSE MIB wih 3-year mauriy (C) An ITM European opion on he FTSE MIB wih 3-year mauriy (D) I canno say Quesion 24. Please read he descripion of he srucured payoff below, draw he payoff graph (assume ha srike is 1% of he value of he underlying a he issuance dae) and explain which opion sraegy replicaes such a payoff. The cerificae guaranees ha a mauriy 1% of he capial invesed is paid back; in addiion, if he value of he underlying a mauriy is lower han 95% of he Srike, hen he cerificaes pays 1% (1 S() Srike ) and if he value of he underlying a mauriy is above 15% of he Srike hen i pays 1% ( S() Srike 1) Answer. This is special case of a double win cerificae similar o he ones analyzed in our lecures in which he payoff funcion is as follows. 2 Payoff 18 16 14 12 1 8 5 1 15 2 8

This srucure may be replicaed by buying a zero coupon bond wih a noional amoun of Eur 1, buying a European pu wih srike equal o 95% of he srike of he srucure, and buying a European call wih srike equal o 15% of he srike of he srucure. The idea is ha he capial saved wih respec o he 1 invesmen in he ZCB should allow o purchase boh he ou-of-he money European pus and calls ha he srucure requires. 9