Indifference fee rate 1

Similar documents
Indifference fee rate for variable annuities

An overview of some financial models using BSDE with enlarged filtrations

M5MF6. Advanced Methods in Derivatives Pricing

Exponential utility maximization under partial information

Forward Dynamic Utility

Utility indifference valuation for non-smooth payoffs on a market with some non tradable assets

Control Improvement for Jump-Diffusion Processes with Applications to Finance

Doubly reflected BSDEs with jumps and generalized Dynkin games

Replication under Price Impact and Martingale Representation Property

Martingale invariance and utility maximization

Structural Models of Credit Risk and Some Applications

Lecture 17. The model is parametrized by the time period, δt, and three fixed constant parameters, v, σ and the riskless rate r.

LECTURE 4: BID AND ASK HEDGING

ON MAXIMIZING DIVIDENDS WITH INVESTMENT AND REINSURANCE

Pricing in markets modeled by general processes with independent increments

Exponential utility maximization under partial information and sufficiency of information

MESURES DE RISQUE DYNAMIQUES DYNAMIC RISK MEASURES

Enlargement of filtration

Robust Portfolio Choice and Indifference Valuation

Optimal Acquisition of a Partially Hedgeable House

RMSC 4005 Stochastic Calculus for Finance and Risk. 1 Exercises. (c) Let X = {X n } n=0 be a {F n }-supermartingale. Show that.

Stochastic Processes and Stochastic Calculus - 9 Complete and Incomplete Market Models

PDE Approach to Credit Derivatives

Arbitrage of the first kind and filtration enlargements in semimartingale financial models. Beatrice Acciaio

Multiple Defaults and Counterparty Risks by Density Approach

Stochastic Differential equations as applied to pricing of options

Ordinary Mixed Life Insurance and Mortality-Linked Insurance Contracts

Variable Annuities with Lifelong Guaranteed Withdrawal Benefits

AMH4 - ADVANCED OPTION PRICING. Contents

SPDE and portfolio choice (joint work with M. Musiela) Princeton University. Thaleia Zariphopoulou The University of Texas at Austin

Girsanov s Theorem. Bernardo D Auria web: July 5, 2017 ICMAT / UC3M

2.1 Mean-variance Analysis: Single-period Model

Risk Measures and Optimal Risk Transfers

1.1 Basic Financial Derivatives: Forward Contracts and Options

Limited liability, or how to prevent slavery in contract theory

Hedging of Contingent Claims under Incomplete Information

Asymmetric information in trading against disorderly liquidation of a large position.

Optimal stopping problems for a Brownian motion with a disorder on a finite interval

Pricing Pension Buy-ins and Buy-outs 1

Deterministic Income under a Stochastic Interest Rate

Option pricing in the stochastic volatility model of Barndorff-Nielsen and Shephard

Polynomial processes in stochastic portofolio theory

Optimal Asset Allocation with Stochastic Interest Rates in Regime-switching Models

Spot and forward dynamic utilities. and their associated pricing systems. Thaleia Zariphopoulou. UT, Austin

Hedging under arbitrage

BACHELIER FINANCE SOCIETY. 4 th World Congress Tokyo, Investments and forward utilities. Thaleia Zariphopoulou The University of Texas at Austin

Non-semimartingales in finance

The Life Cycle Model with Recursive Utility: Defined benefit vs defined contribution.

Insider information and arbitrage profits via enlargements of filtrations

Asset Pricing Models with Underlying Time-varying Lévy Processes

Dynamic Protection for Bayesian Optimal Portfolio

VALUATION OF FLEXIBLE INSURANCE CONTRACTS

Utility Indifference Pricing and Dynamic Programming Algorithm

Hedging with Life and General Insurance Products

The Birth of Financial Bubbles

MATH3075/3975 FINANCIAL MATHEMATICS TUTORIAL PROBLEMS

CONTINUOUS TIME PRICING AND TRADING: A REVIEW, WITH SOME EXTRA PIECES

Hedging under Arbitrage

Replication and Absence of Arbitrage in Non-Semimartingale Models

Drunken Birds, Brownian Motion, and Other Random Fun

Stochastic Dynamical Systems and SDE s. An Informal Introduction

Annuity Decisions with Systematic Longevity Risk. Ralph Stevens

Local Volatility Dynamic Models

CHAPTER 12. Hedging. hedging strategy = replicating strategy. Question : How to find a hedging strategy? In other words, for an attainable contingent

Optimal investments under dynamic performance critria. Lecture IV

MORNING SESSION. Date: Wednesday, April 30, 2014 Time: 8:30 a.m. 11:45 a.m. INSTRUCTIONS TO CANDIDATES

Changes of the filtration and the default event risk premium

Martingale Transport, Skorokhod Embedding and Peacocks

Are stylized facts irrelevant in option-pricing?

Optimal robust bounds for variance options and asymptotically extreme models

On optimal portfolios with derivatives in a regime-switching market

On Using Shadow Prices in Portfolio optimization with Transaction Costs

Optimal Securitization via Impulse Control

Optimal asset allocation under forward performance criteria Oberwolfach, February 2007

Valuation of derivative assets Lecture 8

Economathematics. Problem Sheet 1. Zbigniew Palmowski. Ws 2 dw s = 1 t

Optimal Investment for Worst-Case Crash Scenarios

Multi-period mean variance asset allocation: Is it bad to win the lottery?

Credit Risk in Lévy Libor Modeling: Rating Based Approach

Analytical formulas for local volatility model with stochastic. Mohammed Miri

Time-delayed backward stochastic differential equations: the theory and applications

Exam Quantitative Finance (35V5A1)

Risk minimizing strategies for tracking a stochastic target

Optimal trading strategies under arbitrage

Pricing and Risk Management of guarantees in unit-linked life insurance

Kim Weston (Carnegie Mellon University) Market Stability and Indifference Prices. 1st Eastern Conference on Mathematical Finance.

Are the Azéma-Yor processes truly remarkable?

Risk Neutral Valuation

Personal Finance and Life Insurance under Separation of Risk Aversion and Elasticity of Substitution

The Impact of Natural Hedging on a Life Insurer s Risk Situation

Are the Azéma-Yor processes truly remarkable?

Optimal Execution Beyond Optimal Liquidation

CONVERGENCE OF OPTION REWARDS FOR MARKOV TYPE PRICE PROCESSES MODULATED BY STOCHASTIC INDICES

Valuing volatility and variance swaps for a non-gaussian Ornstein-Uhlenbeck stochastic volatility model

arxiv: v1 [q-fin.pm] 13 Mar 2014

Risk Neutral Measures

All Investors are Risk-averse Expected Utility Maximizers. Carole Bernard (UW), Jit Seng Chen (GGY) and Steven Vanduffel (Vrije Universiteit Brussel)

Valuation of Large Variable Annuity Portfolios: Monte Carlo Simulation and Benchmark Datasets

Investigation of Dependency between Short Rate and Transition Rate on Pension Buy-outs. Arık, A. 1 Yolcu-Okur, Y. 2 Uğur Ö. 2

Credit Risk Models with Filtered Market Information

Transcription:

Indifference fee rate 1 for variable annuities Ricardo ROMO ROMERO Etienne CHEVALIER and Thomas LIM Université d Évry Val d Essonne, Laboratoire de Mathématiques et Modélisation d Evry Second Young researchers meeting on BSDEs, Numerics and Finance. 07-09 July 2014, Bordeaux, France. 1 This research benefitted from the support of the Chaire Marchés en Mutation, Fédération Bancaire Française. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 1 / 31

Outline 1 Variable Annuities 2 Model 3 Indifference fees 4 Numerical Results Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 2 / 31

Outline Variable Annuities 1 Variable Annuities 2 Model 3 Indifference fees 4 Numerical Results Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 3 / 31

Variable Annuities What is a Variable Annuity? Variable annuity is a contract between a policyholder and an insurance company. The policyholder gives an initial amount of money to the insurer. It is invested in a reference portfolio until a preset date, until the policyholder withdraws from the contract or dies. At the end of the contract, the insurance pays an amount of money depending on the performance of the reference portfolio. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 4 / 31

Risks Variable Annuities Actuarial risks: mortality, longevity,.. Financial risks: volatility, interest rate,.. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 5 / 31

Variable Annuities Literature Bauer (2008) presents a general framework to define Variable Annuities (VA). Boyle and Schwartz (1977), extend the Black-Scholes framework to insurance issues. Milvesky and Posner (2001) apply risk neutral option pricing theory to value Guaranteed Minimum Death Benefits (GMDB) in VAs. Dai et al. (2008) HJB equation is derived for a singular control problem related to VA. Belanger et al. (2009) describes the GMDB pricing problem as an impulse control problem. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 6 / 31

Outline Model 1 Variable Annuities 2 Model 3 Indifference fees 4 Numerical Results Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 7 / 31

Model Model (main points) No restrictive assumptions on the reference portfolio and the interest rate dynamics (Markovianity of processes is not assumed): Incomplete market, not a unique risk-neutral measure. We introduce a methodology with BSDEs with a jump. Indifference pricing with continuous fees. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 8 / 31

Model Financial Market and Wealth Process Let (Ω, F, P) be a complete probability space, with F the Brownian filtration. Financial market: ds 0 t = r t S 0 t dt, t [0, T ], S 0 0 = 1, ds t = S t (µ t dt + σ t db t ), t [0, T ], S 0 = s > 0 where µ, σ and r are F-adapted bounded processes and σ is lower bounded by a positive constant. Discounted wealth process: X x,π t = x + t with strategy π and initial capital x. 0 π s (µ s r s )ds + t 0 π s σ s db s, Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 9 / 31

Model Exit time of a Variable Annuity Policy Let τ be the exit time which is the minimum time between: The time of death of the insured. The time of total withdrawal. The random time τ is not assumed to be an F-stopping time. We consider G := (G t ) t 0 with G t := F t σ(1 τ u, u [0, t]) for all t 0. Hypothesis Immersion of F in G: every F-martingale is a G-martingale. The process N. := 1 τ. admits an F-compensator. τ 0 λ t dt. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 10 / 31

Dynamics Model Discounted Account Value A p : da p t = A p t [ (µt r t ξ t p)dt + σ t db t ], t [0, T ], with initial value A 0, fee-rate p and withdrawal (ξ t ) 0 t T. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 11 / 31

Pay-off Model The discounted pay-off including the withdrawals at time T τ to the insured is: F (p) := F L (T, A p )1 {T <τ} + F D (τ, A p )1 {τ T } + T τ 0 ξ s A p s ds. Notice that F (p) is G T τ -measurable. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 12 / 31

Model Guarantees without withdrawals The common guarantees are: Constant guarantee: Gt Q (p) = A 0. Roll-up guarantee: Let be η > 0, then Gt Q (p) = A 0 (1 + η) t. Ratchet guarantee: Gt Q (p) = max(a p t 1, A p t 2,..., A p t ). For t an anniversary date. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 13 / 31

Ricardo ROMO (UniversitéFigure d Evry) : Guarantees Indifferenceand fee rate Account Value. Bordeaux, France 2014 14 / 31 Usual Guarantees Model 2.2 2 Account Process Ratchet Guarantee Roll up Guarantee 1.8 1.6 p 1.4 1.2 1 0.8 0 2 4 6 8 10 12 14 16 18 20 Anniversary Dates

Outline Indifference fees 1 Variable Annuities 2 Model 3 Indifference fees 4 Numerical Results Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 15 / 31

Indifference fees Finding the Indifference Fees The objective is to find a fee p such that sup π A F [0,T ] E [ U ( X x,π T )] = sup E [ U ( A 0 + X x,π T F (p ) )], π A G [0,T ] where A F [0, T ] (resp. A G [0, T ]) is the set of admissible strategies between the interval of time [0, T ] in F (resp. in G). Utility function where γ > 0. U(y) = e γy, y R, Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 16 / 31

Indifference fees The classical problem: V 0 = sup π AF [0,T ] {E [U (X π T )]} Hu, Imkeller and Muller (2004), Rouge and El Karoui (2000) Theorem The optimal value is V 0 = exp(γy 0 ), using the optimal strategy π t := µ t r t γσ 2 t + z t σ t, where y 0 and z are given by the BSDE ( dy t = ν2 t 2γ z tν t ) dt z t db t, y T = 0. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 17 / 31

Indifference fees Utility Maximization with VA (Step 1) V G (p) := sup π AG [0,T ] E [ U ( X A 0,π T F (p) )] Proposition The value function is V G (p) = where F (p) := F (p) + 1 γ log { sup E [ exp ( γ ( X A 0,π T τ F (p) ))], π A G [0,T τ] ess inf π A G [T τ,t ] ) ]} (X E [e γ A 0,π T X A 0,π T τ GT τ. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 18 / 31

Indifference fees Utility Maximization with VA (Step 2) Finding F (p) Proposition There exists a process Y (τ) such that ess inf E [ exp ( (γx A 0,π π A G T X A 0,π T τ )) ] (τ) G T τ = exp(γy T τ ), [T τ,t ] where (Y (τ), Z (τ) ) is solution of the BSDE { ] dy (τ) t = + ν tz (τ) t dt + Z (τ) t db t, [ ν 2 t γ Y (τ) T = 0. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 19 / 31

Indifference fees Utility Maximization with VA (Step 3) V G (p) := sup π AG [0,T ] E [ U ( X A 0,π T F (p) )] Theorem The value function is given by ( V G (p) = exp γ ( A 0 Y 0 (p) )), where (Y (p), Z(p), U(p)) is a solution of Y t (p) = F (p) + T τ t τ T τ Z s (p)db s t τ ( e γus(p) 1 λ s γ T τ t τ ) ν2 s 2γ ν sz s (p) ds U s (p)dh s, t [0, T ]. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 20 / 31

Indifference fees The Optimal Strategy The Strategy πt := ν t γσ t ν t γσ t + Zt(p) σ t, t [0, T τ), + Z (τ) t σ t, t [T τ, T ]. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 21 / 31

Methodology Recapitulation Indifference fees Indifference Fees { [ ( )]} sup π A F [0,T ] {E [U (XT π )]} = sup π A G [0,T ] E U X π,a 0 T F (p). Utility Maximization: Classical Utility Maximization Problem. V0 = exp(γy 0 ). Not the Classical ( Problem. V G (p) = exp γ ( A 0 Y 0 (p) )). Existence of the Indifference Fees. Y 0 (p ) A 0 = y 0. Simulations. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 22 / 31

Indifference fees Existence of the Indifference Fees Consider ψ(p) := Y 0 (p) y 0 A 0, p R. Proposition The function ψ is continuous and non-increasing on R. (i) For any p R, we have ψ(p) > 0 i.e., for any fee p, we have V G (p) < V F. (ii) For any p R, we have ψ(p) < 0 i.e., for any fee p, we have V G (p) > V F. (iii) There exist p 1 and p 2 such that ψ(p 1 )ψ(p 2 ) < 0. Then, there exists an indifference fee p. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 23 / 31

Outline Numerical Results 1 Variable Annuities 2 Model 3 Indifference fees 4 Numerical Results Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 24 / 31

Numerical results. Numerical Results We assume that r and µ are Markov chains taking values in the states spaces S r = {0, 0.01,..., 0.25} and S µ = {0, 0.01, 0.02,..., 0.3}. We give the following numerical values to parameters: γ = 1.3, λ = 0.05, ξ = 0, A 0 = 1, and, for the financial market parameters: r 0 = 0.02, µ 0 = 0.15. Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 25 / 31

Market Risk Numerical Results 0.12 0.1 0.08 p 0.06 0.04 0.02 0 0.1 0.15 0.2 0.25 0.3 0.35 0.4 σ Figure : Ratchet option (T = 20). Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 26 / 31

Actuarial Risk Numerical Results 0.4 0.35 0.3 0.25 p 0.2 0.15 0.1 0.05 0 6 8 10 12 14 16 18 20 22 24 26 28 Anniversary Dates Figure : Ratchet option (σ = 0.3). Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 27 / 31

Numerical Results Roll up Guarantee Risk 0.25 0.2 p 0.15 0.1 0.05 0 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05 η Figure : Roll up option (T = 20, σ = 0.3). Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 28 / 31

The End Thank you! Ricardo ROMO ROMERO rickyromo@gmail.com Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 29 / 31

The End Questions? Ricardo ROMO ROMERO rickyromo@gmail.com Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 30 / 31

The End Their respective transitional matrix are: 1 if i = j, 2 1 if i = 1 and j = 2, 2 qi,j r 1 if i = 27 and j = 26, = 2 1 and if i = j + 1 and i 26, 4 1 if i = j 1 and i 2, 4 0 else, 1 if i = j, 2 1 if i = 1 and j = 2, 2 q µ 1 if i = 32 and j = 31, i,j = 2 1 if i = j + 1 and i 31, 4 1 if i = j 1 and i 2, 4 0 else, Ricardo ROMO (Université d Evry) Indifference fee rate Bordeaux, France 2014 31 / 31