4. Conditional risk measures and their robust representation

Size: px
Start display at page:

Download "4. Conditional risk measures and their robust representation"

Transcription

1 4. Conditional risk measures and their robust representation We consider a discrete-time information structure given by a filtration (F t ) t=0,...,t on our probability space (Ω, F, P ). The time horizon T can be finite or infinite, and we assume F 0 = {, Ω} and F T = F. As the set X of all financial positions we take the space L = L (Ω, F, P ). The subspace L t = L (Ω, F = t, P ) consists of those positions whose outcome only depends on the history up to time t. All inequalities and equalities applied to random variables are meant to hold P -a.s.. Definition. A map ρ t : L L t will be called a monetary conditional risk measure if it satisfies the following properties for all X, Y L : i) Conditional cash invariance: ii) Monotonicity: ρ t (X + X t ) = ρ t (X) X t for any X t L t, X Y ρ t (X) ρ t (Y ), iv) Normalization: ρ t (0) = 0. A conditional risk measure will be called convex if it satisfies iii) Conditional convexity: λ L t, 0 λ 1 ρ t (λx + (1 λ)y ) λρ t (X) + (1 λ)ρ t (Y ). A convex conditional risk measure will be called coherent if it satisfies in addition v) Conditional positive homogeneity: λ L t, λ 0 ρ t (λx) = λρ t (X). For t = 0 the space L t reduces to the real line, and in this case we recover our previous definition of a (unconditional) convex risk measure. To any conditional convex risk measure ρ t we associate its acceptance set (1) A t := {X L ρ t (X) 0}. One easily checks that A t has the following properties: i) conditional convexity, 1

2 ii) X A t, Y X Y A t, iii) ess sup{x L X At } = 0, and 0 A t. Note that ρ t is uniquely determined by its acceptance set since (2) ρ t (X) = ess inf{y L t X + Y At }. A conditional convex risk measure can thus be viewed as the conditional capital requirement needed at time t to make a financial position acceptable at that time. Conversely, one can use acceptance sets to define conditional convex risk measures: If a given acceptance set A t L satisfies the above conditions then the functional ρ t : L L t defined via (2) is a conditional convex risk measure. By M 1 (P ) we denote the set of all probability measures on (Ω, F) which are absolutely continuous with respect to P. As we have shown in section 2, an unconditional convex risk measure which has the Fatou property admits a robust representation of the form (3) ρ(x) = sup (E Q X] α(q)) Q M 1 (P ) with some penalty function α : M 1 (P ) R {+ }. In fact we can take the minimal penalty function (4) α min (Q) = sup E Q X]. X L, ρ ( X) 0 In this section we are going to prove a robust representation for conditional convex risk measures which is analogous to (3). Here the penalty function will depend on the history up to time t, and so it will be given by a map α t from M 1 (P ) to the set of F t -measurable random variables with values in R {+ }. In analogy to (4), we are going to take the minimal penalty function, defined as the worst conditional loss over all positions which are acceptable at time t. To this end we choose versions of the conditional expectations with respect to Q M 1 (P ) which are well defined P - a.s.: (5) E Q X F t ] := Z 1 t E P ( X) Z T F t ] I {Zt >0}, where Z t denotes a Radon-Nikodym density of Q with respect to P on the σ-field F t. In this way, the essential supremum of conditional expected losses, taken over all X A t, is well defined in terms of P. Theorem. For a convex conditional risk measure ρ t the following properties are equivalent: 1. ρ t has the representation (6) ρ t (X) = ess sup Q M1 (P )(E Q X F t ] αt min (Q)), 2

3 where the penalty function α min t is given by (7) α min t (Q) = ess sup X At E Q X F t ]. In fact the representation (6) also holds if we replace M 1 (P ) in (6) by the smaller set P t := {Q M 1 (P ) Q = P on Ft, E P α t (Q)] < }. 2. ρ t has the following Fatou-property: ρ t (X) lim inf n ρ t(x n ) P a.s. for any bounded sequence (X n ) L which converges P a.s. to X L, 3. ρ t is continuous from above, i.e. for any sequence (X n ) L and X L. X n X = ρ t (X n ) ρ t (X) P a.s. Proof. 1) 2): Lebesgue s dominated convergence theorem for conditional expectations, applied to P and to the random variables ( X n )Z T appearing in the version (5) of E Q X n F t ], yields for each Q M 1 (P ). Thus E Q X n F t ] E Q X F t ] P a.s. E Q X F t ] α min t (Q) lim inf n (ess sup Q M1 (P )(E Q X n F t ] αt min (Q)) = lim inf ρ t (X n ) P a.s., and so the representation (6) implies ρ t (X) lim inf ρ t (X n ). 2) 3): The Fatou property yields lim inf ρ t (X n ) ρ t (X). On the other hand we have lim sup ρ t (X n ) ρ t (X) by monotonicity, and this implies the convergence in 3). 3) 1) Since X + ρ t (X) A t for any X L by conditional cash-invariance, the definition of αt min implies hence αt min (Q) E Q X F t ] ρ t (X), ρ t (X) E Q X F t ] αt min (Q) for any Q M 1 (P ). This yields the inequality (8) ρ t (X) ess sup Q M1 (P )(E Q X F t ] αt min (Q)). 3

4 In order to prove the equality in (6), both for M 1 (P ) and for the smaller set P t, it is therefore enough to show that (9) E P ρ t (X)] E P ess sup Q Pt (E Q X F t ] αt min (Q)]). To this end, consider the map ρ : L R defined by ρ(x) := E P ρ t (X)]. It is easy to check that ρ is a convex risk measure. Property 3) implies that ρ is continuous from above; equivalently, it has the Fatou property. Thus ρ has the robust representation ρ(x) = where the penalty function α(q) is given by α(q) = sup (E Q X] α(q)), Q M 1 (P ) sup E Q X]. X L, ρ(x) 0 Next we prove that Q = P on F t if α(q) <. Indeed, take A F t and λ > 0. Then λp A] = E P ρ t (λi A )] = ρ(λi A ) E Q λi A ] α(q), hence P A] QA] + 1 λ α(q) for any λ > 0. Thus α(q) < implies P A] QA] for any A F t, and hence P = Q on F t. Moreover, (10) E P αt min (Q)] α(q) holds for every Q M 1 (P ). Indeed, E P αt min (Q)] = sup E P Y ] by equation (12) below, and so the definition of the penalty function α(q) implies (10) since ρ(x) 0 for all X A t. Thus we have shown that α(q) < implies Q P t. We can now conclude the proof of inequality (8) as follows, using inequality (10) in the third step: E P ρ t (X)] = ρ(x) = sup Q M 1 (P ),α(q)< (E Q X] α(q)) sup Q P t (E Q X] α(q)) sup E P E Q X F t ] αt min (Q)] Q P t E P ess supq Pt (E Q X F t ] α min t (Q) ] ). 4

5 This shows that equality (6) is valid if the essential supremum is only taken over the set P t. In view of inequality (8), this coincides with the essential supremum over all Q M 1 (P ), The following lemma was used in the proof of the Theorem. Lemma. For Q M 1 (P ) and 0 s t, (11) E Q αt min (Q) F s ] = ess sup E Q Y F s ], and in particular (12) E Q αt min (Q)] = sup E Q Y ]. Proof. First we show that the family { EQ X F t ] X At } is directed upward for any Q P t. Indeed, for X, Y A t we can take Z := XI A + Y I A c, where A := {E Q X F t ] E Q Y F t ]} F t. Conditional convexity of ρ t implies Z A t, and clearly we have E Q Z F t ] = max (E Q X F t ], E Q Y F t ]). Thus the family is directed upward. It follows that its essential supremum can be computed as the supremum of an increasing sequence within this family, i.e., there exists a sequence (X Q n ) in A t such that By monotone convergence we get α min t (Q) = lim n E Q X Q n F t ] P a.s. E Q α min t (Q) F s ] = lim n E Q EQ X Q n F t ] Fs ] ess sup E Q Y F s ]. The converse inequality follows directly from the definition of α min t (Q). This shows (11), and (12) can be viewed as the special case s = 0. Example: The entropic conditional risk measure. Suppose that preferences are characterized by an exponential utility function u(x) = 1 exp( βx) with β > 0. At time t the conditional expected utility of a financial position X L is then given by the F t - measurable random variable U t (X) = E1 e βx F t ]. 5

6 The set A t := { X L Ut (X) U t (0) } = { X L Ee βx F t ] 1 } satisfies the necessary conditions for an acceptance set. The induced convex conditional risk measure ρ t is the conditional entropic measure given by ρ t (X) = ess inf { Y L t = ess inf { Y L t } X + Y A t Ee βx F t ] e βy }, i.e., ρ t (X) = 1 γ log Ee γx F t ]. It is easy to see that ρ t has the Fatou property and thus admits a robust representation of the form (6). In analogy to the unconditional case, the minimal penalty function takes the form α t (Q) = 1 β Ĥt(Q P ), where Ĥt(Q P ) denotes the conditional relative entropy of Q with respect to P, given F t : Ĥ t (Q P ) := E Q log Z ] T Ft Z t = E P ZT Z t log Z T Z t F t ] I {Zt >0}; note that Z t > 0 Q-a.s., and that Z t = 1 for Q P t. This is the reason why ρ t is called the conditional entropic risk measure. In the coherent case we obtain the following representation result: Corollary. Let ρ t be a coherent conditional risk measure which has the Fatou property. Then ρ t has the representation (13) ρ t (X) = ess sup Q Q t E Q X F t ], where Q t := { Q M 1 (P ) Q = P on Ft, α min t (Q) = 0 P a.s. }. Proof. In the coherent case the penalty function αt min (Q) can only take the values 0 or. Indeed, for A := {αt min (Q) > 0}, X A t and any λ > 0 we have λi A X A t due to conditional positive homogeneity of ρ t, hence 6

7 α min t (Q) = ess sup X At E Q X F t ] ess sup X At E Q λi A X F t ] = λi A αt min (Q), and the lower bound converges to on A as λ. Thus αt min (Q) = on A. But for Q P t we have E P αt min (Q)] <, and this implies P A] = 0. Thus P t coincides with Q t, and the representation (6), where we are free to take P t instead of M 1 (P ), reduces to (13). Example: Conditional Value at Risk. For any α (0, 1), the acceptance set A t = {X L P X < 0 F t ] α} defines a monetary conditional risk measure, called conditional value at risk at level α: α (X) Ft = ess inf{m t L t P X + m t < 0 F t ] α} = ess sup{c t L t P X < c t F t ] α} = q α (X) Ft, where q α (X) Ft is a conditional quantile of X, i.e., a quantile of the conditional distribution of X under P, given the σ-field F t. Conditional Value at Risk is conditionally positively homogenous, but it is not conditionally convex. However, we do obtain a coherent conditional risk measure by taking the average up to some level λ (0, 1]. The resulting risk measure α (X) Ft := 1 λ λ 0 α (X)dα is called conditional Value at Risk at level λ. In analogy to the unconditional case it can be shown that the following representation holds: α (X) Ft = ess sup{e Q X F t ] Q = P on F t, dq dp 1 λ }. Remark. The representations (6) and (13) can be formulated in terms of equivalent probability measures Q P if ρ t is sensitive with respect to P in the sense that P ρ t ( ɛi A ) > 0] > 0 for all ɛ > 0 and for any A F such that P A] > 0. More precisely, let M e 1(P ) denote the set of all probability measures on (Ω, F) which are equivalent to P, and let ρ t be sensitive with respect to P. Then we can replace M 1 (P ) by M e 1(P ) in (6), in the definition of P t in the Theorem, and in the definition of Q t in the Corollary. 7

Risk Measures in non-dominated Models

Risk Measures in non-dominated Models Purpose: Study Risk Measures taking into account the model uncertainty in mathematical finance. Plan 1 Non-dominated Models Model Uncertainty Fundamental topological Properties 2 Risk Measures on L p (c)

More information

Dynamic risk measures. Robust representation and examples

Dynamic risk measures. Robust representation and examples VU University Amsterdam Faculty of Sciences Jagiellonian University Faculty of Mathematics and Computer Science Joint Master of Science Programme Dorota Kopycka Student no. 1824821 VU, WMiI/69/04/05 UJ

More information

Set-Valued Risk Measures and Bellman s Principle

Set-Valued Risk Measures and Bellman s Principle Set-Valued Risk Measures and Bellman s Principle Zach Feinstein Electrical and Systems Engineering, Washington University in St. Louis Joint work with Birgit Rudloff (Vienna University of Economics and

More information

Entropic risk measures: coherence vs. convexity, model ambiguity, and robust large deviations

Entropic risk measures: coherence vs. convexity, model ambiguity, and robust large deviations Entropic risk measures: coherence vs. convexity, model ambiguity, robust large deviations Hans Föllmer Thomas Knispel 2 Abstract We study a coherent version of the entropic risk measure, both in the lawinvariant

More information

RISK MEASURES ON ORLICZ HEART SPACES

RISK MEASURES ON ORLICZ HEART SPACES Communications on Stochastic Analysis Vol. 9, No. 2 (2015) 169-180 Serials Publications www.serialspublications.com RISK MEASURES ON ORLICZ HEART SPACES COENRAAD LABUSCHAGNE, HABIB OUERDIANE, AND IMEN

More information

Robust Return Risk Measures

Robust Return Risk Measures Noname manuscript No. will be inserted by the editor) Robust Return Risk Measures Fabio Bellini Roger J. A. Laeven Emanuela Rosazza Gianin Received: date / Accepted: date Abstract In this paper we provide

More information

A Note on Robust Representations of Law-Invariant Quasiconvex Functions

A Note on Robust Representations of Law-Invariant Quasiconvex Functions A Note on Robust Representations of Law-Invariant Quasiconvex Functions Samuel Drapeau Michael Kupper Ranja Reda October 6, 21 We give robust representations of law-invariant monotone quasiconvex functions.

More information

Time Consistency in Decision Making

Time Consistency in Decision Making Time Consistency in Decision Making Igor Cialenco Department of Applied Mathematics Illinois Institute of Technology cialenco@iit.edu Joint work with Tomasz R. Bielecki and Marcin Pitera Mathematical Finance

More information

Aggregate Risk. MFM Practitioner Module: Quantitative Risk Management. John Dodson. February 6, Aggregate Risk. John Dodson.

Aggregate Risk. MFM Practitioner Module: Quantitative Risk Management. John Dodson. February 6, Aggregate Risk. John Dodson. MFM Practitioner Module: Quantitative Risk Management February 6, 2019 As we discussed last semester, the general goal of risk measurement is to come up with a single metric that can be used to make financial

More information

Lectures for the Course on Foundations of Mathematical Finance

Lectures for the Course on Foundations of Mathematical Finance Definitions and properties of Lectures for the Course on Foundations of Mathematical Finance First Part: Convex Marco Frittelli Milano University The Fields Institute, Toronto, April 2010 Definitions and

More information

Convex Risk Measures: Basic Facts, Law-invariance and beyond, Asymptotics for Large Portfolios

Convex Risk Measures: Basic Facts, Law-invariance and beyond, Asymptotics for Large Portfolios Convex Risk Measures: Basic Facts, Law-invariance and beyond, Asymptotics for Large Portfolios Hans Föllmer a Thomas Knispel b Abstract This paper provides an introduction to the theory of capital requirements

More information

Generalized quantiles as risk measures

Generalized quantiles as risk measures Generalized quantiles as risk measures Bellini, Klar, Muller, Rosazza Gianin December 1, 2014 Vorisek Jan Introduction Quantiles q α of a random variable X can be defined as the minimizers of a piecewise

More information

Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets

Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets Tim Leung 1, Qingshuo Song 2, and Jie Yang 3 1 Columbia University, New York, USA; leung@ieor.columbia.edu 2 City

More information

On convex risk measures on L p -spaces

On convex risk measures on L p -spaces MMOR manuscript No. (will be inserted by the editor) On convex risk measures on L p -spaces M. Kaina and L. Rüschendorf University of Freiburg, Department of Mathematical Stochastics, Eckerstr. 1, 79104

More information

Thomas Knispel Leibniz Universität Hannover

Thomas Knispel Leibniz Universität Hannover Optimal long term investment under model ambiguity Optimal long term investment under model ambiguity homas Knispel Leibniz Universität Hannover knispel@stochastik.uni-hannover.de AnStAp0 Vienna, July

More information

Conditional and Dynamic Preferences

Conditional and Dynamic Preferences Conditional and Dynamic Preferences How can we Understand Risk in a Dynamic Setting? Samuel Drapeau Joint work with Hans Föllmer Humboldt University Berlin Jena - March 17th 2009 Samuel Drapeau Joint work

More information

Coherent risk measures

Coherent risk measures Coherent risk measures Foivos Xanthos Ryerson University, Department of Mathematics Toµɛας Mαθηµατ ικὼν, E.M.Π, 11 Noɛµβρὶoυ 2015 Research interests Financial Mathematics, Mathematical Economics, Functional

More information

Lecture 3: Expected Value. These integrals are taken over all of Ω. If we wish to integrate over a measurable subset A Ω, we will write

Lecture 3: Expected Value. These integrals are taken over all of Ω. If we wish to integrate over a measurable subset A Ω, we will write Lecture 3: Expected Value 1.) Definitions. If X 0 is a random variable on (Ω, F, P), then we define its expected value to be EX = XdP. Notice that this quantity may be. For general X, we say that EX exists

More information

The Axiomatic Approach to Risk Measures for Capital Determination

The Axiomatic Approach to Risk Measures for Capital Determination The Axiomatic Approach to Risk Measures for Capital Determination Hans Föllmer Humboldt-Universität zu Berlin Stefan Weber Leibniz Universität Hannover January 5, 2015 1 Introduction The quantification

More information

Coherent and convex monetary risk measures for unbounded

Coherent and convex monetary risk measures for unbounded Coherent and convex monetary risk measures for unbounded càdlàg processes Patrick Cheridito ORFE Princeton University Princeton, NJ, USA dito@princeton.edu Freddy Delbaen Departement für Mathematik ETH

More information

Coherent Risk Measures. Acceptance Sets. L = {X G : X(ω) < 0, ω Ω}.

Coherent Risk Measures. Acceptance Sets. L = {X G : X(ω) < 0, ω Ω}. So far in this course we have used several different mathematical expressions to quantify risk, without a deeper discussion of their properties. Coherent Risk Measures Lecture 11, Optimisation in Finance

More information

Performance Measures for Ranking and Selection Procedures

Performance Measures for Ranking and Selection Procedures Rolf Waeber Performance Measures for Ranking and Selection Procedures 1/23 Performance Measures for Ranking and Selection Procedures Rolf Waeber Peter I. Frazier Shane G. Henderson Operations Research

More information

Merging of Opinions under Uncertainty

Merging of Opinions under Uncertainty Working Papers Institute of Mathematical Economics 433 Mai 2010 Merging of Opinions under Uncertainty Monika Bier and Daniel Engelage IMW Bielefeld University Postfach 100131 33501 Bielefeld Germany email:

More information

A unified approach to time consistency of dynamic risk measures and dynamic performance measures in discrete time

A unified approach to time consistency of dynamic risk measures and dynamic performance measures in discrete time A unified approach to time consistency of dynamic risk measures and dynamic performance measures in discrete time Tomasz R. Bielecki a, Igor Cialenco a, and Marcin Pitera b First Circulated: September

More information

Strongly Consistent Multivariate Conditional Risk Measures

Strongly Consistent Multivariate Conditional Risk Measures Strongly Consistent Multivariate Conditional Risk Measures annes offmann Thilo Meyer-Brandis regor Svindland January 11, 2016 Abstract We consider families of strongly consistent multivariate conditional

More information

Distribution-Invariant Risk Measures, Entropy, and Large Deviations

Distribution-Invariant Risk Measures, Entropy, and Large Deviations Distribution-Invariant Risk Measures, Entropy, and Large Deviations Stefan Weber Cornell University June 24, 2004; this version December 4, 2006 Abstract The simulation of distributions of financial positions

More information

Monetary Risk Measures and Generalized Prices Relevant to Set-Valued Risk Measures

Monetary Risk Measures and Generalized Prices Relevant to Set-Valued Risk Measures Applied Mathematical Sciences, Vol. 8, 2014, no. 109, 5439-5447 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2014.43176 Monetary Risk Measures and Generalized Prices Relevant to Set-Valued

More information

Theory of Probability Fall 2008

Theory of Probability Fall 2008 MIT OpenCourseWare http://ocw.mit.edu 8.75 Theory of Probability Fall 008 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. Section Prekopa-Leindler inequality,

More information

Finanzrisiken. Fachbereich Angewandte Mathematik - Stochastik Introduction to Financial Risk Measurement

Finanzrisiken. Fachbereich Angewandte Mathematik - Stochastik Introduction to Financial Risk Measurement ment Convex ment 1 Bergische Universität Wuppertal, Fachbereich Angewandte Mathematik - Stochastik @math.uni-wuppertal.de Inhaltsverzeichnis ment Convex Convex Introduction ment Convex We begin with the

More information

Journal of Mathematical Economics. Coherent risk measures in general economic models and price bubbles

Journal of Mathematical Economics. Coherent risk measures in general economic models and price bubbles Journal of Mathematical Economics ( ) Contents lists available at SciVerse ScienceDirect Journal of Mathematical Economics journal homepage: www.elsevier.com/locate/jmateco Coherent risk measures in general

More information

Coherent and convex monetary risk measures for bounded

Coherent and convex monetary risk measures for bounded Coherent and convex monetary risk measures for bounded càdlàg processes Patrick Cheridito ORFE Princeton University Princeton, NJ, USA dito@princeton.edu Freddy Delbaen Departement für Mathematik ETH Zürich

More information

Comparative and qualitative robustness for law-invariant risk measures

Comparative and qualitative robustness for law-invariant risk measures Comparative and qualitative robustness for law-invariant risk measures Volker Krätschmer Alexander Schied Henryk Zähle Abstract When estimating the risk of a P&L from historical data or Monte Carlo simulation,

More information

Regulatory Arbitrage of Risk Measures

Regulatory Arbitrage of Risk Measures Regulatory Arbitrage of Risk Measures Ruodu Wang November 29, 2015 Abstract We introduce regulatory arbitrage of risk measures as one of the key considerations in choosing a suitable risk measure to use

More information

Robust preferences and robust portfolio choice

Robust preferences and robust portfolio choice Robust preferences and robust portfolio choice Hans FÖLLMER Institut für Mathematik Humboldt-Universität Unter den Linden 6 10099 Berlin, Germany foellmer@math.hu-berlin.de Alexander SCHIED School of ORIE

More information

Pricing, Hedging and Optimally Designing Derivatives via Minimization of Risk Measures 1

Pricing, Hedging and Optimally Designing Derivatives via Minimization of Risk Measures 1 Pricing, Hedging and Optimally Designing Derivatives via Minimization of Risk Measures 1 11th July 2005 Pauline Barrieu 2 and Nicole El Karoui 3 The question of pricing and hedging a given contingent claim

More information

An axiomatic characterization of capital allocations of coherent risk measures

An axiomatic characterization of capital allocations of coherent risk measures An axiomatic characterization of capital allocations of coherent risk measures Michael Kalkbrener Deutsche Bank AG Abstract An axiomatic definition of coherent capital allocations is given. It is shown

More information

CVaR and Examples of Deviation Risk Measures

CVaR and Examples of Deviation Risk Measures CVaR and Examples of Deviation Risk Measures Jakub Černý Department of Probability and Mathematical Statistics Stochastic Modelling in Economics and Finance November 10, 2014 1 / 25 Contents CVaR - Dual

More information

Lecture 22 Girsanov s Theorem

Lecture 22 Girsanov s Theorem Lecture 22: Girsanov s Theorem of 8 Course: Theory of Probability II Term: Spring 25 Instructor: Gordan Zitkovic Lecture 22 Girsanov s Theorem An example Consider a finite Gaussian random walk X n = n

More information

Brownian Motion. 1 Definition Brownian Motion Wiener measure... 3

Brownian Motion. 1 Definition Brownian Motion Wiener measure... 3 Brownian Motion Contents 1 Definition 2 1.1 Brownian Motion................................. 2 1.2 Wiener measure.................................. 3 2 Construction 4 2.1 Gaussian process.................................

More information

Representation theorem for AVaR under a submodular capacity

Representation theorem for AVaR under a submodular capacity 3 214 5 ( ) Journal of East China Normal University (Natural Science) No. 3 May 214 Article ID: 1-5641(214)3-23-7 Representation theorem for AVaR under a submodular capacity TIAN De-jian, JIANG Long, JI

More information

Minimax and risk averse multistage stochastic programming

Minimax and risk averse multistage stochastic programming Minimax and risk averse multistage stochastic programming Alexander Shapiro School of Industrial & Systems Engineering, Georgia Institute of Technology, 765 Ferst Drive, Atlanta, GA 30332. Abstract. In

More information

An inverse of Sanov s theorem

An inverse of Sanov s theorem An inverse of Sanov s theorem Ayalvadi Ganesh and Neil O Connell BRIMS, Hewlett-Packard Labs, Bristol Abstract Let X k be a sequence of iid random variables taking values in a finite set, and consider

More information

Lecture 21: Expectation of CRVs, Fatou s Lemma and DCT Integration of Continuous Random Variables

Lecture 21: Expectation of CRVs, Fatou s Lemma and DCT Integration of Continuous Random Variables EE50: Probability Foundations for Electrical Engineers July-November 205 Lecture 2: Expectation of CRVs, Fatou s Lemma and DCT Lecturer: Krishna Jagannathan Scribe: Jainam Doshi In the present lecture,

More information

MATH MEASURE THEORY AND FOURIER ANALYSIS. Contents

MATH MEASURE THEORY AND FOURIER ANALYSIS. Contents MATH 3969 - MEASURE THEORY AND FOURIER ANALYSIS ANDREW TULLOCH Contents 1. Measure Theory 2 1.1. Properties of Measures 3 1.2. Constructing σ-algebras and measures 3 1.3. Properties of the Lebesgue measure

More information

Lecture 2: Random Variables and Expectation

Lecture 2: Random Variables and Expectation Econ 514: Probability and Statistics Lecture 2: Random Variables and Expectation Definition of function: Given sets X and Y, a function f with domain X and image Y is a rule that assigns to every x X one

More information

J. Banasiak Department of Mathematics and Applied Mathematics University of Pretoria, Pretoria, South Africa BANACH LATTICES IN APPLICATIONS

J. Banasiak Department of Mathematics and Applied Mathematics University of Pretoria, Pretoria, South Africa BANACH LATTICES IN APPLICATIONS J. Banasiak Department of Mathematics and Applied Mathematics University of Pretoria, Pretoria, South Africa BANACH LATTICES IN APPLICATIONS Contents 1 Introduction.........................................................................

More information

MATHS 730 FC Lecture Notes March 5, Introduction

MATHS 730 FC Lecture Notes March 5, Introduction 1 INTRODUCTION MATHS 730 FC Lecture Notes March 5, 2014 1 Introduction Definition. If A, B are sets and there exists a bijection A B, they have the same cardinality, which we write as A, #A. If there exists

More information

LECTURE 15: COMPLETENESS AND CONVEXITY

LECTURE 15: COMPLETENESS AND CONVEXITY LECTURE 15: COMPLETENESS AND CONVEXITY 1. The Hopf-Rinow Theorem Recall that a Riemannian manifold (M, g) is called geodesically complete if the maximal defining interval of any geodesic is R. On the other

More information

Decision principles derived from risk measures

Decision principles derived from risk measures Decision principles derived from risk measures Marc Goovaerts Marc.Goovaerts@econ.kuleuven.ac.be Katholieke Universiteit Leuven Decision principles derived from risk measures - Marc Goovaerts p. 1/17 Back

More information

2 Statement of the problem and assumptions

2 Statement of the problem and assumptions Mathematical Notes, 25, vol. 78, no. 4, pp. 466 48. Existence Theorem for Optimal Control Problems on an Infinite Time Interval A.V. Dmitruk and N.V. Kuz kina We consider an optimal control problem on

More information

A NOTE ON JENSEN S INEQUALITY INVOLVING MONETARY UTILITY FUNCTIONS

A NOTE ON JENSEN S INEQUALITY INVOLVING MONETARY UTILITY FUNCTIONS Electronic Journal of Mathematical Analysis and Applications Vol. 6(2) July 2018, pp. 68-75. ISSN: 2090-729X(online) http://fcag-egypt.com/journals/ejmaa/ A NOTE ON JENSEN S INEQUALITY INVOLVING MONETARY

More information

PROBLEMS. (b) (Polarization Identity) Show that in any inner product space

PROBLEMS. (b) (Polarization Identity) Show that in any inner product space 1 Professor Carl Cowen Math 54600 Fall 09 PROBLEMS 1. (Geometry in Inner Product Spaces) (a) (Parallelogram Law) Show that in any inner product space x + y 2 + x y 2 = 2( x 2 + y 2 ). (b) (Polarization

More information

1 Stochastic Dynamic Programming

1 Stochastic Dynamic Programming 1 Stochastic Dynamic Programming Formally, a stochastic dynamic program has the same components as a deterministic one; the only modification is to the state transition equation. When events in the future

More information

Class Notes for Math 921/922: Real Analysis, Instructor Mikil Foss

Class Notes for Math 921/922: Real Analysis, Instructor Mikil Foss University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln Math Department: Class Notes and Learning Materials Mathematics, Department of 200 Class Notes for Math 92/922: Real Analysis,

More information

Key words. quasiconvex functions, dual representation, quasiconvex optimization, dynamic risk measures, conditional certainty equivalent.

Key words. quasiconvex functions, dual representation, quasiconvex optimization, dynamic risk measures, conditional certainty equivalent. DUAL REPRESENTATION OF QUASICONVEX CONDITIONAL MAPS MARCO FRITTELLI AND MARCO MAGGIS Abstract. We provide a dual representation of quasiconvex maps π : L F L G, between two locally convex lattices of random

More information

Markov Chain BSDEs and risk averse networks

Markov Chain BSDEs and risk averse networks Markov Chain BSDEs and risk averse networks Samuel N. Cohen Mathematical Institute, University of Oxford (Based on joint work with Ying Hu, Robert Elliott, Lukas Szpruch) 2nd Young Researchers in BSDEs

More information

1: PROBABILITY REVIEW

1: PROBABILITY REVIEW 1: PROBABILITY REVIEW Marek Rutkowski School of Mathematics and Statistics University of Sydney Semester 2, 2016 M. Rutkowski (USydney) Slides 1: Probability Review 1 / 56 Outline We will review the following

More information

Optimal Risk Sharing with Different Reference Probabilities

Optimal Risk Sharing with Different Reference Probabilities Optimal Risk Sharing with Different Reference Probabilities Beatrice Acciaio Gregor Svindland July 29, 2008 Abstract We investigate the problem of optimal risk sharing between agents endowed with cashinvariant

More information

On Kusuoka Representation of Law Invariant Risk Measures

On Kusuoka Representation of Law Invariant Risk Measures MATHEMATICS OF OPERATIONS RESEARCH Vol. 38, No. 1, February 213, pp. 142 152 ISSN 364-765X (print) ISSN 1526-5471 (online) http://dx.doi.org/1.1287/moor.112.563 213 INFORMS On Kusuoka Representation of

More information

MATH 418: Lectures on Conditional Expectation

MATH 418: Lectures on Conditional Expectation MATH 418: Lectures on Conditional Expectation Instructor: r. Ed Perkins, Notes taken by Adrian She Conditional expectation is one of the most useful tools of probability. The Radon-Nikodym theorem enables

More information

Problem set 1, Real Analysis I, Spring, 2015.

Problem set 1, Real Analysis I, Spring, 2015. Problem set 1, Real Analysis I, Spring, 015. (1) Let f n : D R be a sequence of functions with domain D R n. Recall that f n f uniformly if and only if for all ɛ > 0, there is an N = N(ɛ) so that if n

More information

3 (Due ). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure?

3 (Due ). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure? MA 645-4A (Real Analysis), Dr. Chernov Homework assignment 1 (Due ). Show that the open disk x 2 + y 2 < 1 is a countable union of planar elementary sets. Show that the closed disk x 2 + y 2 1 is a countable

More information

Optimal stopping for non-linear expectations Part I

Optimal stopping for non-linear expectations Part I Stochastic Processes and their Applications 121 (2011) 185 211 www.elsevier.com/locate/spa Optimal stopping for non-linear expectations Part I Erhan Bayraktar, Song Yao Department of Mathematics, University

More information

The Canonical Model Space for Law-invariant Convex Risk Measures is L 1

The Canonical Model Space for Law-invariant Convex Risk Measures is L 1 The Canonical Model Space for Law-invariant Convex Risk Measures is L 1 Damir Filipović Gregor Svindland 3 November 2008 Abstract In this paper we establish a one-to-one correspondence between lawinvariant

More information

Regularly Varying Asymptotics for Tail Risk

Regularly Varying Asymptotics for Tail Risk Regularly Varying Asymptotics for Tail Risk Haijun Li Department of Mathematics Washington State University Humboldt Univ-Berlin Haijun Li Regularly Varying Asymptotics for Tail Risk Humboldt Univ-Berlin

More information

Advanced Probability

Advanced Probability Advanced Probability Perla Sousi October 10, 2011 Contents 1 Conditional expectation 1 1.1 Discrete case.................................. 3 1.2 Existence and uniqueness............................ 3 1

More information

Applications of Ito s Formula

Applications of Ito s Formula CHAPTER 4 Applications of Ito s Formula In this chapter, we discuss several basic theorems in stochastic analysis. Their proofs are good examples of applications of Itô s formula. 1. Lévy s martingale

More information

Robust Optimal Control Using Conditional Risk Mappings in Infinite Horizon

Robust Optimal Control Using Conditional Risk Mappings in Infinite Horizon Robust Optimal Control Using Conditional Risk Mappings in Infinite Horizon Kerem Uğurlu Monday 9 th April, 2018 Department of Applied Mathematics, University of Washington, Seattle, WA 98195 e-mail: keremu@uw.edu

More information

Nonlinear stabilization via a linear observability

Nonlinear stabilization via a linear observability via a linear observability Kaïs Ammari Department of Mathematics University of Monastir Joint work with Fathia Alabau-Boussouira Collocated feedback stabilization Outline 1 Introduction and main result

More information

Letting p shows that {B t } t 0. Definition 0.5. For λ R let δ λ : A (V ) A (V ) be defined by. 1 = g (symmetric), and. 3. g

Letting p shows that {B t } t 0. Definition 0.5. For λ R let δ λ : A (V ) A (V ) be defined by. 1 = g (symmetric), and. 3. g 4 Contents.1 Lie group p variation results Suppose G, d) is a group equipped with a left invariant metric, i.e. Let a := d e, a), then d ca, cb) = d a, b) for all a, b, c G. d a, b) = d e, a 1 b ) = a

More information

Pareto Optimal Allocations for Law Invariant Robust Utilities

Pareto Optimal Allocations for Law Invariant Robust Utilities Pareto Optimal Allocations for Law Invariant Robust Utilities on L 1 Claudia Ravanelli Swiss Finance Institute Gregor Svindland University of Munich and EPFL November 2012 Abstract We prove the existence

More information

Problem Set 1: Solutions Math 201A: Fall Problem 1. Let (X, d) be a metric space. (a) Prove the reverse triangle inequality: for every x, y, z X

Problem Set 1: Solutions Math 201A: Fall Problem 1. Let (X, d) be a metric space. (a) Prove the reverse triangle inequality: for every x, y, z X Problem Set 1: s Math 201A: Fall 2016 Problem 1. Let (X, d) be a metric space. (a) Prove the reverse triangle inequality: for every x, y, z X d(x, y) d(x, z) d(z, y). (b) Prove that if x n x and y n y

More information

Risk-Averse Dynamic Optimization. Andrzej Ruszczyński. Research supported by the NSF award CMMI

Risk-Averse Dynamic Optimization. Andrzej Ruszczyński. Research supported by the NSF award CMMI Research supported by the NSF award CMMI-0965689 Outline 1 Risk-Averse Preferences 2 Coherent Risk Measures 3 Dynamic Risk Measurement 4 Markov Risk Measures 5 Risk-Averse Control Problems 6 Solution Methods

More information

Lecture 2. We now introduce some fundamental tools in martingale theory, which are useful in controlling the fluctuation of martingales.

Lecture 2. We now introduce some fundamental tools in martingale theory, which are useful in controlling the fluctuation of martingales. Lecture 2 1 Martingales We now introduce some fundamental tools in martingale theory, which are useful in controlling the fluctuation of martingales. 1.1 Doob s inequality We have the following maximal

More information

Preparatory Material for the European Intensive Program in Bydgoszcz 2011 Analytical and computer assisted methods in mathematical models

Preparatory Material for the European Intensive Program in Bydgoszcz 2011 Analytical and computer assisted methods in mathematical models Preparatory Material for the European Intensive Program in Bydgoszcz 2011 Analytical and computer assisted methods in mathematical models September 4{18 Basics on the Lebesgue integral and the divergence

More information

Uniformly Uniformly-ergodic Markov chains and BSDEs

Uniformly Uniformly-ergodic Markov chains and BSDEs Uniformly Uniformly-ergodic Markov chains and BSDEs Samuel N. Cohen Mathematical Institute, University of Oxford (Based on joint work with Ying Hu, Robert Elliott, Lukas Szpruch) Centre Henri Lebesgue,

More information

The Subdifferential of Convex Deviation Measures and Risk Functions

The Subdifferential of Convex Deviation Measures and Risk Functions The Subdifferential of Convex Deviation Measures and Risk Functions Nicole Lorenz Gert Wanka In this paper we give subdifferential formulas of some convex deviation measures using their conjugate functions

More information

Robust control and applications in economic theory

Robust control and applications in economic theory Robust control and applications in economic theory In honour of Professor Emeritus Grigoris Kalogeropoulos on the occasion of his retirement A. N. Yannacopoulos Department of Statistics AUEB 24 May 2013

More information

Dice -sion Making under Uncertainty: When Can a Random Decision Reduce Risk?

Dice -sion Making under Uncertainty: When Can a Random Decision Reduce Risk? Dice -sion Making under Uncertainty: When Can a Random Decision Reduce Risk? Erick Delage 1, Daniel Kuhn 2, and Wolfram Wiesemann 3 1 Department of Decision Sciences, HEC Montréal, Canada 2 Risk Analytics

More information

Iowa State University. Instructor: Alex Roitershtein Summer Homework #1. Solutions

Iowa State University. Instructor: Alex Roitershtein Summer Homework #1. Solutions Math 501 Iowa State University Introduction to Real Analysis Department of Mathematics Instructor: Alex Roitershtein Summer 015 EXERCISES FROM CHAPTER 1 Homework #1 Solutions The following version of the

More information

Convex risk measures on L p

Convex risk measures on L p Convex risk measures on L p Damir Filipović Gregor Svindland Mathematics Institute University of Munich 8333 Munich, Germany 6 July 27 Abstract Convex risk measures are best known on L. In this paper we

More information

Risk-Consistent Conditional Systemic Risk Measures

Risk-Consistent Conditional Systemic Risk Measures Risk-Consistent Conditional Systemic Risk Measures Thilo Meyer-Brandis University of Munich joint with: H. Hoffmann and G. Svindland, University of Munich Workshop on Systemic Risk and Financial Networks

More information

Risk Preferences and their Robust Representation

Risk Preferences and their Robust Representation Risk Preferences and their Robust Representation Samuel Drapeau Michael Kupper July 212 To address the plurality of interpretations of the subjective notion of risk, we describe it by means of a risk order

More information

Financial Asset Price Bubbles under Model Uncertainty

Financial Asset Price Bubbles under Model Uncertainty Financial Asset Price Bubbles under Model Uncertainty Francesca Biagini, Jacopo Mancin October 24, 27 Abstract We study the concept of financial bubbles in a market model endowed with a set P of probability

More information

A Note on the Swiss Solvency Test Risk Measure

A Note on the Swiss Solvency Test Risk Measure A Note on the Swiss Solvency Test Risk Measure Damir Filipović and Nicolas Vogelpoth Vienna Institute of Finance Nordbergstrasse 15 A-1090 Vienna, Austria first version: 16 August 2006, this version: 16

More information

Part II Probability and Measure

Part II Probability and Measure Part II Probability and Measure Theorems Based on lectures by J. Miller Notes taken by Dexter Chua Michaelmas 2016 These notes are not endorsed by the lecturers, and I have modified them (often significantly)

More information

Probability and Measure

Probability and Measure Part II Year 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2018 84 Paper 4, Section II 26J Let (X, A) be a measurable space. Let T : X X be a measurable map, and µ a probability

More information

3 Integration and Expectation

3 Integration and Expectation 3 Integration and Expectation 3.1 Construction of the Lebesgue Integral Let (, F, µ) be a measure space (not necessarily a probability space). Our objective will be to define the Lebesgue integral R fdµ

More information

Dynamic Assessment Indices

Dynamic Assessment Indices Dynamic Assessment Indices Tomasz R. Bielecki Department of Applied Mathematics, Illinois Institute of Technology, Chicago, 60616 IL, USA bielecki@iit.edu Samuel Drapeau Humboldt-Universität Berlin, Unter

More information

Multistage stochastic programs: Time-consistency, time-inconsistency and martingale bounds

Multistage stochastic programs: Time-consistency, time-inconsistency and martingale bounds Multistage stochastic programs: Time-consistency, time-inconsistency and martingale bounds Georg Ch. Pflug, joint work with Raimund Kovacevic and Alois Pichler April 2016 The time-consistency problem Many

More information

(1) Consider the space S consisting of all continuous real-valued functions on the closed interval [0, 1]. For f, g S, define

(1) Consider the space S consisting of all continuous real-valued functions on the closed interval [0, 1]. For f, g S, define Homework, Real Analysis I, Fall, 2010. (1) Consider the space S consisting of all continuous real-valued functions on the closed interval [0, 1]. For f, g S, define ρ(f, g) = 1 0 f(x) g(x) dx. Show that

More information

g 2 (x) (1/3)M 1 = (1/3)(2/3)M.

g 2 (x) (1/3)M 1 = (1/3)(2/3)M. COMPACTNESS If C R n is closed and bounded, then by B-W it is sequentially compact: any sequence of points in C has a subsequence converging to a point in C Conversely, any sequentially compact C R n is

More information

Operations Research Letters. On a time consistency concept in risk averse multistage stochastic programming

Operations Research Letters. On a time consistency concept in risk averse multistage stochastic programming Operations Research Letters 37 2009 143 147 Contents lists available at ScienceDirect Operations Research Letters journal homepage: www.elsevier.com/locate/orl On a time consistency concept in risk averse

More information

A gentle introduction to imprecise probability models

A gentle introduction to imprecise probability models A gentle introduction to imprecise probability models and their behavioural interpretation Gert de Cooman gert.decooman@ugent.be SYSTeMS research group, Ghent University A gentle introduction to imprecise

More information

Introduction and Preliminaries

Introduction and Preliminaries Chapter 1 Introduction and Preliminaries This chapter serves two purposes. The first purpose is to prepare the readers for the more systematic development in later chapters of methods of real analysis

More information

Stochastic dominance with respect to a capacity and risk measures

Stochastic dominance with respect to a capacity and risk measures Stochastic dominance with respect to a capacity and risk measures Miryana Grigorova To cite this version: Miryana Grigorova. Stochastic dominance with respect to a capacity and risk measures. 2011.

More information

Random Process Lecture 1. Fundamentals of Probability

Random Process Lecture 1. Fundamentals of Probability Random Process Lecture 1. Fundamentals of Probability Husheng Li Min Kao Department of Electrical Engineering and Computer Science University of Tennessee, Knoxville Spring, 2016 1/43 Outline 2/43 1 Syllabus

More information

Deterministic Dynamic Programming

Deterministic Dynamic Programming Deterministic Dynamic Programming 1 Value Function Consider the following optimal control problem in Mayer s form: V (t 0, x 0 ) = inf u U J(t 1, x(t 1 )) (1) subject to ẋ(t) = f(t, x(t), u(t)), x(t 0

More information

Point Process Control

Point Process Control Point Process Control The following note is based on Chapters I, II and VII in Brémaud s book Point Processes and Queues (1981). 1 Basic Definitions Consider some probability space (Ω, F, P). A real-valued

More information

On maxitive integration

On maxitive integration On maxitive integration Marco E. G. V. Cattaneo Department of Mathematics, University of Hull m.cattaneo@hull.ac.uk Abstract A functional is said to be maxitive if it commutes with the (pointwise supremum

More information