Leader-Follower Stochastic Differential Game with Asymmetric Information and Applications

Size: px
Start display at page:

Download "Leader-Follower Stochastic Differential Game with Asymmetric Information and Applications"

Transcription

1 Leader-Follower Stochastic Differential Game with Asymmetric Information and Applications Jie Xiong Department of Mathematics University of Macau Macau Based on joint works with Shi and Wang 2015 Barrett Lecture, Knoxville, 14/05/2015]

2 Outline 1 Motivation 2 Problem formulation 3 LQ Leader-Follower Stochastic Differential Game with Asymmetric Information 4 Follower s optimization problem 5 Complete Information LQ Stochastic Optimal Control Problem of the Leader

3 1. A motivation Example: Continuous Time Newsvendor Problems D( ) is the demand rate { dd(t) = a(µ kr(t) D(t))dt + σdw (t) + σd W (t), D(0) = d 0 R, retail price R(t).

4 The retailer will obtain an expected profit J 1 (q( ), R( ), w( )) T [ = E (R(t) S) min[d(t), q(t)] 0 ] (w(t) S)q(t) dt. order rate q(t); wholesale price w(t); salvaged at unit price S 0.

5 the manufacturer has a fixed production cost per unit M 0, an expected profit J 2 (q( ), R( ), w( )) = E T 0 (w(t) M)q(t)dt.

6 The information G 1 t, G 2 t available to the retailer and the manufacturer at time t, respectively. G 1 t G 2 t. For any w( ), retailer selects a G 1 t -adapted processes pair (q ( ), R ( )) for the retailer such that J 1 (q ( ), R ( ), w( )) J 1 ( q ( ; w( )), R ( ; w( )), w( ) ) = max q( ),R( ) J 1(q( ), R( ), w( )),

7 Manufacturer selects a G 2 t -adapted process w ( ) for the manufacturer such that J 2 (q ( ), R ( ), w ( )) ( J 2 q ( ; w ( )), R ( ; w ( )), w ( ) ) = max J ( 2 q ( ; w( )), R ( ; w( )), w( ) ), w( )

8 Example: Cooperative Advertising and Pricing Problems dx(t) = [ ρu(t) 1 x(t) δx(t) ] dt + σ(x(t))dw (t) + σ(x(t))d W (t), x(0) = x 0 [0, 1], u(t) advertisement effort rate, x(t) awareness share, ρ response constant, and δ the rate at which potential consumers are lost. Manufacturer decide: wholesale price w(t), cooperative participation rate θ(t). Retailer decides: effort u(t), retail price p(t).

9 Given w(t) and θ(t), retailer solves an optimization problem to maximize his expected profit = E J 1 (w( ), θ( ), u( ), p( )) T D is demand function. 0 e rt[ (p(t) w(t))d(p(t))x(t) ] (1 θ(t))u 2 (t) dt,

10 The manufacturer s optimization problem is to maximize his expected profit J 2 (w( ), θ( ), u( ), p( )) T = E e rt[ ] (w(t) c)d(p(t))x(t) θ(t)u 2 (t) dt, 0 where c 0 is the constant unit production cost.

11 The information G 1 t, G 2 t available to the retailer and the manufacturer at time t, respectively. G 1 t G 2 t. Given w and θ, retailer chooses G 1,t -adapted pair (u ( ), p ( )), J 1 (w( ), θ( ), u ( ), p ( )) J 1 ( w( ), θ( ), u ( ; (w( ), θ( )), p ( ; (w( ), θ( )) ) = max u( ),p( ) 0 J 1(w( ), θ( ), u( ), p( )).

12 Select a G 2,t -adapted pair (w ( ), θ ( )) for the manufacturer J 2 (w ( ), θ ( ), u ( ), p ( )) ( J 2 w ( ), θ ( ), u ( ; (w ( ), θ ( )), p ( ; (w ( ), θ ( )) ) = max J ( 2 w( ), θ( ), u ( ; (w( ), θ( )), w( ),0 θ( ) 1 p ( ; (w( ), θ( )) ),

13 2. Problem formulation State equation: dx u 1,u 2 (t) = b ( t, x u 1,u 2 (t), u 1 (t), u 2 (t) ) dt + σ ( t, x u 1,u 2 (t), u 1 (t), u 2 (t) ) dw (t) + σ ( t, x u 1,u 2 (t), u 1 (t), u 2 (t) ) d W (t), x u 1,u 2 (0) = x 0, (2.1)

14 Gt 1 Gt 2 F t. The admissible control { U 1 := u u1 1 : Ω [0, T ] U 1 is Gt 1 -adapted and sup E u 1 (t) i <, i = 1, 2, 0 t T { U 2 := u 2 u 2 : Ω [0, T ] U 2 is Gt 2 -adapted and sup E u 2 (t) i <, i = 1, 2, 0 t T }, }. (2.2) (2.3)

15 Problem of the follower. For any chosen u 2 ( ) U 2 by the leader, choose a G 1 t -adapted control u 1 ( ) = u 1 ( ; u 2( )) U 1, such that J 1 (u 1( ), u 2 ( )) J 1 ( u 1 ( ; u 2 ( )), u 2 ( ) ) = inf u 1 U 1 J 1 (u 1 ( ), u 2 ( )), u 1 ( ) = u 1 ( ; u 2( )), (partial information) optimal control. (2.4)

16 The leader would like to choose a Gt 2 -adapted control u 2 ( ) to minimize J 2 (u 1( ), u 2 ( )) [ T ( = E g 2 t, x u 1,u 2 (t), u 1(t; u 2 (t)), u 2 (t) ) dt 0 ] + G 2 (x u 1,u 2 (T )). (2.5)

17 Problem of the leader. Find a G 2 t -adapted control u 2 ( ) U 2, such that J 2 (u 1( ), u 2( )) = J 2 ( u 1 ( ; u 2( )), u 2( ) ) = inf u 2 U 2 J 2 ( u 1 ( ; u 2 ( )), u 2 ( ) ), (2.6)

18 Issacs (1954-5) differential games Basar and Olsder (1982) Monograph to summarize Stackelberg (1934) Leader-follower game introduced Yong (2002) LQ, complete info. Bensoussan et al (2012) Stoch. Max. Principle

19 3. LQ Leader-Follower Stochastic Differential Game with Asymmetric Information; Stochastic maximum principle under partial information. The state equation dx u 1,u 2 (t) = [ A(t)x u 1,u 2 (t) + B 1 (t)u 1 (t) + B 2 (t)u 2 (t) ] dt + [ C(t)x u 1,u 2 (t) + D 1 (t)u 1 (t) + D 2 (t)u 2 (t) ] dw (t) + [ u C(t)x 1,u 2 (t) + D 1 (t)u 1 (t) + D 2 (t)u 2 (t) ] d W (t), x u 1,u 2 (0) = x 0, (3.7)

20 Given u 2 U 2, follower minimizes his cost functional J 1 (u 1 ( ), u 2 ( )) = 1 [ T ( Q1 2 E (t)x u 1,u 2 (t), x u 1,u 2 (t) 0 + N 1 (t)u 1 (t), u 1 (t) ) dt (3.8) + G 1 x u 1,u 2 (T ), x u 1,u 2 (T ) ], Info: G 1 t = σ{ W (s); 0 s t}

21 Stochastic maximum principle under partial information. Probability space (Ω, F, ˆP). State equation: FBSDE dx = b(t, x, v)dt + σ(t, x, v)dw + σ(t, x, v)d W, dy = g(t, x, y, z, z, v)dt zdw zdy, x(0) = x 0, y(1) = f(x(1)). (3.9) Observation { dy = h(t, x)dt + d W, Y (0) = 0. (3.10)

22 Admissible controls v A v is G t -adapted, where G t = F Y t. Control problem: Find u A s.t. where ( 1 J(v) = Ê 0 J(u) = min v A J(v), ) l(t, x v, y v, z v, z v, v)dt + φ(x v (1)) + γ(y v (0)).

23 Hypothesis (H1): Coeff. conti. diff. w/ bounded 1st order partial derivatives. σ, h bounded. Hypothesis (H2): l, φ, γ conti. diff. ( 1 ) Ê l(t, x v, y v, z v, z v, v) dt + φ(x v (1)) + γ(y v (0)) <. 0

24 when b = σ = σ = 0, studied by Huang-Wang-Xiong (SICON 2009) General form by Wang-Wu-Xiong (SICON 2013). Detail on LQ case by Wang-Wu-Xiong (IEEE TAC 2015).

25 Replace b, W in state equation by b, Y, resp, where b = b σh. State equation: dx = b(t, x, v)dt + σ(t, x, v)dw + σ(t, x, v)dy, dy = g(t, x, y, z, z, v)dt zdw zdy, x(0) = x 0, y(1) = f(x(1)). (3.11)

26 Let So { t Z v (t) = exp h(s, x v )dy t 0 } h 2 (s, x v )ds. dz v = Z v h(t, x v )dy. (3.12) Let P ˆP be s.t. dˆp = Z v (1)dP. Then, under P, (W, Y ) is a B.M. ( 1 ) J(v) = E Z v l(t, x v, y v, z v, z v, v)dt + Z v (1)φ(x v (1)) + γ(y v (0)). 0

27 Hypothesis (H3): For ξ bdd, G t -measurable, then v A. v(s) ξ1 [t,t+τ) (s), For any u A and v being G s -adapted and bdd, δ > 0, if ɛ < δ, then u + ɛv A.

28 Adjoint equations: dp = (g y (t, x, y, z, z, u)p l y (t, x, y, z, z, u)) dt + (g z (t, x, y, z, z, u)p l z (t, x, y, z, z, u)) dw { + (g z (t, x, y, z, z, u)p h(t, x)) d W, dq = (b x (t, x, u) σ(t, x, u)h x (t, x))q + σ x (t, x, u)k + σ x (t, x, u) k g x } (t, x, y, z, z, u)p l x (t, x, y, z, z, u) dt kdw kd W, p(0) = γ y (y(0)) q(1) = f x (x(1))p(1) + φ x (x(1)), (3.13)

29 and { dp = l(t, x, y, z, z, u)dt QdW Qd W P (1) = φ(x(1)). (3.14) Hamitonian H(t, x, y, z, z, v; p, q, k, k, Q) = bq + σk + σ k + h Q + l (g h z)p.

30 Theorem If u is a local minimum for J( ), then ( ) E H u (t, x, y, z, z, v; p, q, k, k, Q) G t = 0. Idea of proof Set d dɛ J(u + ɛv) ɛ=0 = 0.

31 Remark (3.14) is for Z v (t). If h = 0, this eq. is not needed. Suppose h = b = σ = σ, it is proved in Huang-Wang-Xiong (2008, SICON)

32 4. Follower s optimization problem Step 1. (Optimal control) Hamiltonian function: H 1 ( t, x, u1, u 2 ; q, k, k ) = q, A(t)x + B 1 (t)u 1 + B 2 (t)u 2 + k, C(t)x + D 1 (t)u 1 + D 2 (t)u 2 + k, C(t)x + D1 (t)u 1 + D 2 (t)u 2 (4.15) 1 Q1 (t)x, x 1 N1 (t)u 1, u

33 Then, 0 = N 1 (t)u 1(t) B1 (t)e [ q(t) ] G 1 t D1 (t)e [ k(t) G t 1 ] D 1 (t)e [ k(t) (4.16) Gt 1 ], a.e.t [0, T ], where F t -adapted (q( ), k( ), k( )) satisfies the BSDE [ dq(t) = A (t)q(t) + C (t)k(t) + C (t) k(t) ] Q 1 (t)x u 1,u 2 (t) dt k(t)dw (t) k(t)d W (t), q(t ) = G 1 x u 1,u 2 (T ). (4.17)

34 Step 2. (Optimal filtering) Let ˆf(t) := E [ f(t) G t 1 ], f = q, k, k. How to derive filtering equation? Direct from (4.17) is impossible. Solve (4.17) first! Guess: q(t) = P 1 (t)x u 1,u 2 (t) ϕ(t), t [0, T ], (4.18) Set { dϕ(t) = α(t)dt + β(t)d W (t), ϕ(t ) = 0. (4.19)

35 Applying dq(t) and comparing, we get [ ] k(t) = P 1 (t) C(t)x u 1,u 2 (t) + D 1 (t)u 1(t) + D 2 (t)u 2 (t), (4.20) k(t) = P1 (t)[ C(t)x u 1,u 2 (t) + D 1 (t)u 1(t) + D ] 2 (t)u 2 (t) β(t), [ α(t) = P 1 (t) P 1 (t)a(t) A (t)p 1 (t) ] Q 1 (t) x u 1,u 2 (t) P 1 (t)b(t)u 1(t) P 1 (t)b 2 (t)u 2 (t) A (t)ϕ(t) + C (t)k(t) + C (t) k(t). (4.21) (4.22)

36 Hence, ˆq(t) = P 1 (t)ˆx u 1,û 2 (t) ϕ(t), (4.23) [ ] ˆk(t) = P 1 (t) C(t)ˆx u 1,û 2 (t) + D 1 (t)u 1(t) + D 2 (t)û 2 (t), (4.24) [ ˆ k(t) = P 1 (t) C(t)ˆx u 1,û 2 (t) + D 1 (t)u 1(t) + D ] 2 (t)û 2 (t) β(t) (4.25) with û 2 (t) := E [ u 2 (t) G 1,t ].

37 Using filtering technique (Lemma 5.4 in Xiong (2008)), we get dˆx u 1,û 2 (t) = [ A(t)ˆx u 1,û 2 (t) + B 1 (t)u 1(t) + B 2 (t)û 2 (t) ] dt + [ u C(t)ˆx 1,û 2 (t) + D 1 (t)u 1(t) + D 2 (t)û 2 (t) ] (4.26) d W (t), ˆx u 1,û 2 (0) = x 0, and { dˆq(t) = A (t)ˆq(t) + C (t)ˆk(t) + C (t)ˆ k(t) } Q 1 (t)ˆx u 1,û 2 (t) dt ˆ k(t)d W (t), ˆq(T ) = G 1ˆx u 1,û 2 (T ), (4.27)

38 Step 3. (Optimal state feedback control) By (4.16) we arrive at UNIVERSIDADE DE MACAU [ ( u 1(t) = Ñ1(t) 1 B1 (t)p 1 (t) + D1 (t)p 1 (t)c(t) + D 1 (t)p 1 (t) C(t) ) ˆx u 1,û 2 (t) ( + D1 (t)p 1 (t)d 2 (t) + D 1 (t)p 1 (t) D ) 2 (t) û 2 (t) ] + B1 (t)ϕ(t) + D 1 (t)β(t), (4.28) where Ñ 1 (t) := N 1 (t) + D 1 (t)p 1 (t)d 1 (t) + D 1 (t)p 1 (t) D 1 (t) (4.29)

39 Substituting (4.28) into (4.22), we obtain Riccati equation P 1 (t) + P 1 (t)a(t) + A (t)p 1 (t) + C (t)p 1 (t)c(t) + C (t)p 1 (t) C(t) + Q 1 (t) ( P 1 (t)b 1 (t) + C (t)p 1 (t)d 1 (t) )Ñ 1 + C (t)p 1 (t) D 1 (t) 1 (B (t) 1 (t)p 1 (t) + D1 (t)p 1 (t)c(t) + D 1 (t)p 1 (t) C(t) ) = 0, P 1 (T ) = G 1. Suppose it admits a unique differentiable solution P 1 ( ) (4.30)

40 [ α(t) = A (t) S ] 1 (t)ñ1(t) 1 B1 (t) ϕ(t) [ + C (t) S ] 1 (t)ñ1(t) 1 D 1 (t) β(t) [ + S ] 1 (t)ñ1(t) 1 S(t) + S2 (t) û 2 (t), (4.31) which is Gt 1 -adapted, where S(t) := D1 (t)p 1 (t)d 2 (t) + D 1 (t)p 1 (t) D 2 (t), S 1 (t) := P 1 (t)b 1 (t) + C (t)p 1 (t)d 1 (t) + C (t)p 1 (t) D 1 (t), S 2 (t) := P 1 (t)b 2 (t) + C (t)p 1 (t)d 2 (t) + C (t)p 1 (t) D 2 (t), t [0, T ].

41 Then, { [ dϕ(t) = A (t) S ] 1 (t)ñ1(t) 1 B1 (t) ϕ(t) [ + C (t) S ] 1 (t)ñ1(t) 1 D 1 (t) β(t) [ + S ] } 1 (t)ñ1(t) 1 S(t) + S2 (t) û 2 (t) dt (4.32) ϕ(t ) = 0, β(t)d W (t), which admits a unique G 1 t -adapted solution (ϕ( ), β( )).

42 Finally, dˆx u 1,û 2 u (t) = [Ã(t)ˆx 1,û 2 (t) + F 1 (t)ϕ(t) + B 1 (t)β(t) + B ] 2 (t)û 2 (t) dt [ u + C(t)ˆx 1,û 2 (t) + B 1 (t)ϕ(t) + F ] 3 (t)β(t) + D2 (t)û 2 (t) d W (t), (4.33) ˆx u 1,û 2 (0) = x 0, solves ˆx u 1,û 2, where

43 Ã(t) := A(t) B 1 (t)ñ1(t) 1 S 1 (t), B 2 (t) := B 2 (t) B 1 (t)ñ1(t) 1 S(t), F 1 (t) := B 1 (t)ñ1(t) 1 B 1 (t), B 1 (t) := B 1 (t)ñ1(t) 1 D 1 (t), C(t) := C(t) D 1 (t)ñ1(t) 1 S 1 (t), F 3 (t) := D 1 (t)ñ1(t) 1 D 1 (t), D 2 (t) := D 2 (t) D 1 (t)ñ1(t) 1 S(t) F 4 (t) := S 1 (t)ñ1(t) 1 S(t) + S2 (t).

44 Theorem 3.1 Let Riccati equation admit a differentiable solution P 1 ( ) and let matrix be convertible. For chosen u 2 ( ) of the leader, Problem of the follower admits an optimal control u 1 ( ) of the state feedback, where processes triple (ˆx u 1,û 2 ( ), ϕ( ), β( )) is the unique Gt 1 -adapted solution to the FBSDE (4.32)-(4.33).

45 5. Complete Information LQ Stochastic Optimal Control Problem of the Leader The leader knows the complete information F t at any time t. The state equations faced by the leader consist of BSDE (4.32) and the SDE (3.7). By (4.28), it can be written as

46 [ dx u 2 (t) = A(t)x u 2 (t) + ( Ã(t) A(t) )ˆxû2 (t) + F 1 (t)ϕ(t) + B 1 (t)β(t) + B 2 (t)u 2 (t) + ( B2 (t) B 2 (t) ) ] û 2 (t) dt [ + C(t)x u 2 (t) + F 5 (t)ˆxû2 (t) + B1 (t) ϕ(t) + D1 (t)β(t) + D 2 (t)u 2 (t) + F ] 2 (t)û 2 (t) dw (t) [ + C(t)x u 2 (t) + ( C(t) û C(t) )ˆx 2 (t) + B 1 (t) ϕ(t) + F 3 (t)β(t) + D 2 (t)u 2 (t) + ( ] D(t) D(t) )û2 (t) d W (t), dϕ(t) = [ Ã (t)ϕ(t) + C (t)β(t) + F 4 (t)û 2 (t) ] dt β(t)d W (t), x u 2 (0) = x 0, ϕ(t ) = 0. (5.34)

47 The leader s cost functional J 2 (u 2 ( )) := J 2 (u 1( ), u 2 ( )) = 1 [ T ( Q2 2 E (t)x u 1,u 2 (t), x u 1,u 2 (t) 0 + N 2 (t)u 2 (t), u 2 (t) ) dt + G 2 x u 1,u 2 (T ), x u 1,u 2 (T ) ] (5.35) This complete info conditional mean-field LQ problem with state (5.34) and cost (5.34) can be solved explicitly using a general result.

48 General conditional mean-field SMP. State dx u 2 (t) = b(t)dt + σ(t)dw (t) + σ(t)d W (t), { d 1 dq(t) = b x (t)q(t) + σx(t)k j j (t) d 2 + j=1 j=1 } σ x(t) k j j (t) g 1x (t) dt k(t)dw (t) k(t)d W (t), x u 2 (0) = x 0, q(t ) = G 1x (x u 2 (T )), (5.36) where b is a function of (t, x, ˆx, u 2, û 2, q, k, k, ˆk, ˆ k).

49 Cost: [ T ] J 2 (u 2 ( )) = E g 2 (t)dt + G 2 (x u 2 (T )) 0 Problem of the leader. Find a Gt 2 -adapted control u 2 ( ) U 2, such that J 2 (u 2( )) = inf J 2 (u 2 ( )), (5.37) u 2 U 2 subject to (5.36).

50 Define Hamiltonian H 2 ( t, x, u2, ˆx, û 2, q, k, k; y, z, z, p ) = y, b() + tr { z σ(t) } + tr { z σ(t) } + g 2 (t) p, b x (t)q + d 1 j=1 d 2 σx(t)k j j + j=1 σ x(t) k j j g 1x (t). (5.38)

51 Let (y( ), z( ), z( ), p( )) be the unique F t -adapted solution to the adjoint conditional mean-field FBSDE of the leader { dp(t) = b x(t)p(t) + E [ b ˆx (t)p(t) ] } G 1 t dt d 1 { + σx j (t)p(t) + E [ σ j ˆx (t)p(t) G t 1 ] } dw j (t) j=1 d 2 { + σ x j j (t)p(t) + E [ σ ˆx (t)p(t) G t 1 ] } d W j (t), j=1 (5.39)

52 dy = j=1 UNIVERSIDADE DE MACAU { b x y + E [ bˆx y G t 1 ] d 1 [ + σxz j j + E [ σ jˆx zj G t 1 ] ] d 2 + [ σ x(t)z j j (t) + E [ σ jˆx (t)zj (t) ] ] G 1 t j=1 n i=1 { bx [ bx (t)q(t)p i (t) + E (t)q(t)p i (t) x i ˆx i n { i=1 x i n { i=1 x i ( d1 j=1 ( d1 j=1 + g 1xx p + E [ g 1xˆx p G 1 t [ ( d1 σxk )p j j i + E ˆx i j=1 [ ( d1 σ x k )p j j i + E ˆx i j=1 G 1 t ]} σ j xk j ) p i G 1 t ]} σ j x k j ) p i G 1 t ]} ] + g2x + E [ g 2ˆx Gt 1 ] } dt zdw zd W,

53 with boundary p(0) = 0, y(t ) = G 1xx (x (T ))p(t ) + G 2x (x (T )), (5.41) where we have used φ φ ( t, x (t), ˆx (t), u 2 (t), û 2 (t)) for φ = b, σ, σ, g 1, g 2 and all their derivatives. Related work: Anderson and Djehiche (2011), Li (2012), Yong (2.13).

54 Proposition (Maximum principle of conditional mean-field FBSDE with partial information) Let u 2 ( ) U 2 be the optimal control for Problem of the leader and (x ( ), q ( ), k ( ), k ( )) be the corresponding optimal state which is the unique F t -adapted solution to (5.36). Let (y( ), z( ), z( ), p( )) be the unique F t -adapted solution to the adjoint equation. Then { [ ] } H 2 H2 E + E u 2 û 2 G1 t G2 t = 0, a.e.t [0, T ], (5.42) u 2 =u 2 (t)

55 Now, we come back to LQ problem. Define Hamiltonian H 2 ( t, x u 2, u 2, ϕ, β; y, z, z, p ) = y, A(t)x u 2 + ( Ã(t) A(t) )ˆxû2 + F 1 (t)ϕ + B 1 (t)β + B 2 (t)u 2 + ( B2 (t) B 2 (t) ) û 2 + z, C(t)x u 2 + F 5 (t)ˆxû2 + B 1 (t)ϕ + D1 (t)β + D 2 (t)u 2 + F 2 (t)u 2 + z, C(t)x u 2 + ( C(t) û C(t) )ˆx 2 + B 1 (t)ϕ + F 3 (t)β + D 2 (t)u 2 + ( D2 (t) D 2 (t) ) û 2 + 1[ p, Ã ϕ + C β + F4 û 2 + Q2 x u 2, x u 2 + ] N 2 u 2, u 2. 2 (5.43)

56 Applying the general thm, if there exists an F t -adapted optimal control u 2 ( ) U 2 for the leader, then 0 = N 2 (t)u 2(t) + F 4 (t)ˆp(t) + B 2 (t)y(t) + ( B2 (t) B 2 (t) ) ŷ(t) + D 2 (t)z(t) + F 2 (t)ẑ(t) + D 2 (t) z(t) (5.44) + ( D2 (t) D 2 (t) ) ˆ z(t), a.e. t [0, T ], where F t -adapted processes quadruple (p( ), y( ), z( ), z( )) satisfies the conditional mean-field FBSDE

57 dp(t) = [ Ã(t)p(t) + F 1 (t)y(t) + B1 (t)z(t) + B 1 (t) z(t) ] dt + [ C(t)p(t) + B 1 (t)y(t) + D 1 (t)z(t) + F 3 (t) z(t) ] d W (t), dy(t) = [ A (t)y(t) + ( Ã(t) A(t) ) ŷ(t) + C (t)z(t) + F 5 (t)ẑ(t) + C (t) z(t) + ( C(t) C(t) ) ˆ z(t) + Q2 (t)x (t) ] dt z(t)dw (t) z(t)d W (t), p(0) = 0, y(t ) = G 2 x (T ). (5.45)

58 Thanks!

HJB equations. Seminar in Stochastic Modelling in Economics and Finance January 10, 2011

HJB equations. Seminar in Stochastic Modelling in Economics and Finance January 10, 2011 Department of Probability and Mathematical Statistics Faculty of Mathematics and Physics, Charles University in Prague petrasek@karlin.mff.cuni.cz Seminar in Stochastic Modelling in Economics and Finance

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

Nonlinear representation, backward SDEs, and application to the Principal-Agent problem

Nonlinear representation, backward SDEs, and application to the Principal-Agent problem Nonlinear representation, backward SDEs, and application to the Principal-Agent problem Ecole Polytechnique, France April 4, 218 Outline The Principal-Agent problem Formulation 1 The Principal-Agent problem

More information

Mean-field SDE driven by a fractional BM. A related stochastic control problem

Mean-field SDE driven by a fractional BM. A related stochastic control problem Mean-field SDE driven by a fractional BM. A related stochastic control problem Rainer Buckdahn, Université de Bretagne Occidentale, Brest Durham Symposium on Stochastic Analysis, July 1th to July 2th,

More information

Lecture 4: Ito s Stochastic Calculus and SDE. Seung Yeal Ha Dept of Mathematical Sciences Seoul National University

Lecture 4: Ito s Stochastic Calculus and SDE. Seung Yeal Ha Dept of Mathematical Sciences Seoul National University Lecture 4: Ito s Stochastic Calculus and SDE Seung Yeal Ha Dept of Mathematical Sciences Seoul National University 1 Preliminaries What is Calculus? Integral, Differentiation. Differentiation 2 Integral

More information

Mean Field Game Theory for Systems with Major and Minor Agents: Nonlinear Theory and Mean Field Estimation

Mean Field Game Theory for Systems with Major and Minor Agents: Nonlinear Theory and Mean Field Estimation Mean Field Game Theory for Systems with Major and Minor Agents: Nonlinear Theory and Mean Field Estimation Peter E. Caines McGill University, Montreal, Canada 2 nd Workshop on Mean-Field Games and Related

More information

On continuous time contract theory

On continuous time contract theory Ecole Polytechnique, France Journée de rentrée du CMAP, 3 octobre, 218 Outline 1 2 Semimartingale measures on the canonical space Random horizon 2nd order backward SDEs (Static) Principal-Agent Problem

More information

LECTURES ON MEAN FIELD GAMES: II. CALCULUS OVER WASSERSTEIN SPACE, CONTROL OF MCKEAN-VLASOV DYNAMICS, AND THE MASTER EQUATION

LECTURES ON MEAN FIELD GAMES: II. CALCULUS OVER WASSERSTEIN SPACE, CONTROL OF MCKEAN-VLASOV DYNAMICS, AND THE MASTER EQUATION LECTURES ON MEAN FIELD GAMES: II. CALCULUS OVER WASSERSTEIN SPACE, CONTROL OF MCKEAN-VLASOV DYNAMICS, AND THE MASTER EQUATION René Carmona Department of Operations Research & Financial Engineering PACM

More information

Optimal Control of Predictive Mean-Field Equations and Applications to Finance

Optimal Control of Predictive Mean-Field Equations and Applications to Finance Optimal Control of Predictive Mean-Field Equations and Applications to Finance Bernt Øksendal and Agnès Sulem Abstract We study a coupled system of controlled stochastic differential equations (SDEs) driven

More information

An Application of Pontryagin s Maximum Principle in a Linear Quadratic Differential Game

An Application of Pontryagin s Maximum Principle in a Linear Quadratic Differential Game An Application of Pontryagin s Maximum Principle in a Linear Quadratic Differential Game Marzieh Khakestari (Corresponding author) Institute For Mathematical Research, Universiti Putra Malaysia, 43400

More information

Non-zero Sum Stochastic Differential Games of Fully Coupled Forward-Backward Stochastic Systems

Non-zero Sum Stochastic Differential Games of Fully Coupled Forward-Backward Stochastic Systems Non-zero Sum Stochastic Differential Games of Fully Coupled Forward-Backward Stochastic Systems Maoning ang 1 Qingxin Meng 1 Yongzheng Sun 2 arxiv:11.236v1 math.oc 12 Oct 21 Abstract In this paper, an

More information

= m(0) + 4e 2 ( 3e 2 ) 2e 2, 1 (2k + k 2 ) dt. m(0) = u + R 1 B T P x 2 R dt. u + R 1 B T P y 2 R dt +

= m(0) + 4e 2 ( 3e 2 ) 2e 2, 1 (2k + k 2 ) dt. m(0) = u + R 1 B T P x 2 R dt. u + R 1 B T P y 2 R dt + ECE 553, Spring 8 Posted: May nd, 8 Problem Set #7 Solution Solutions: 1. The optimal controller is still the one given in the solution to the Problem 6 in Homework #5: u (x, t) = p(t)x k(t), t. The minimum

More information

Optimal investment strategies for an index-linked insurance payment process with stochastic intensity

Optimal investment strategies for an index-linked insurance payment process with stochastic intensity for an index-linked insurance payment process with stochastic intensity Warsaw School of Economics Division of Probabilistic Methods Probability space ( Ω, F, P ) Filtration F = (F(t)) 0 t T satisfies

More information

Lecture 12: Diffusion Processes and Stochastic Differential Equations

Lecture 12: Diffusion Processes and Stochastic Differential Equations Lecture 12: Diffusion Processes and Stochastic Differential Equations 1. Diffusion Processes 1.1 Definition of a diffusion process 1.2 Examples 2. Stochastic Differential Equations SDE) 2.1 Stochastic

More information

MEAN FIELD GAMES WITH MAJOR AND MINOR PLAYERS

MEAN FIELD GAMES WITH MAJOR AND MINOR PLAYERS MEAN FIELD GAMES WITH MAJOR AND MINOR PLAYERS René Carmona Department of Operations Research & Financial Engineering PACM Princeton University CEMRACS - Luminy, July 17, 217 MFG WITH MAJOR AND MINOR PLAYERS

More information

Risk-Sensitive and Robust Mean Field Games

Risk-Sensitive and Robust Mean Field Games Risk-Sensitive and Robust Mean Field Games Tamer Başar Coordinated Science Laboratory Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign Urbana, IL - 6181 IPAM

More information

Generalized Riccati Equations Arising in Stochastic Games

Generalized Riccati Equations Arising in Stochastic Games Generalized Riccati Equations Arising in Stochastic Games Michael McAsey a a Department of Mathematics Bradley University Peoria IL 61625 USA mcasey@bradley.edu Libin Mou b b Department of Mathematics

More information

Stochastic Differential Equations

Stochastic Differential Equations CHAPTER 1 Stochastic Differential Equations Consider a stochastic process X t satisfying dx t = bt, X t,w t dt + σt, X t,w t dw t. 1.1 Question. 1 Can we obtain the existence and uniqueness theorem for

More information

Solution of Stochastic Optimal Control Problems and Financial Applications

Solution of Stochastic Optimal Control Problems and Financial Applications Journal of Mathematical Extension Vol. 11, No. 4, (2017), 27-44 ISSN: 1735-8299 URL: http://www.ijmex.com Solution of Stochastic Optimal Control Problems and Financial Applications 2 Mat B. Kafash 1 Faculty

More information

arxiv: v1 [math.pr] 18 Oct 2016

arxiv: v1 [math.pr] 18 Oct 2016 An Alternative Approach to Mean Field Game with Major and Minor Players, and Applications to Herders Impacts arxiv:161.544v1 [math.pr 18 Oct 216 Rene Carmona August 7, 218 Abstract Peiqi Wang The goal

More information

Harnack Inequalities and Applications for Stochastic Equations

Harnack Inequalities and Applications for Stochastic Equations p. 1/32 Harnack Inequalities and Applications for Stochastic Equations PhD Thesis Defense Shun-Xiang Ouyang Under the Supervision of Prof. Michael Röckner & Prof. Feng-Yu Wang March 6, 29 p. 2/32 Outline

More information

Large Deviations Principles for McKean-Vlasov SDEs, Skeletons, Supports and the law of iterated logarithm

Large Deviations Principles for McKean-Vlasov SDEs, Skeletons, Supports and the law of iterated logarithm Large Deviations Principles for McKean-Vlasov SDEs, Skeletons, Supports and the law of iterated logarithm Gonçalo dos Reis University of Edinburgh (UK) & CMA/FCT/UNL (PT) jointly with: W. Salkeld, U. of

More information

SDE Coefficients. March 4, 2008

SDE Coefficients. March 4, 2008 SDE Coefficients March 4, 2008 The following is a summary of GARD sections 3.3 and 6., mainly as an overview of the two main approaches to creating a SDE model. Stochastic Differential Equations (SDE)

More information

The Wiener Itô Chaos Expansion

The Wiener Itô Chaos Expansion 1 The Wiener Itô Chaos Expansion The celebrated Wiener Itô chaos expansion is fundamental in stochastic analysis. In particular, it plays a crucial role in the Malliavin calculus as it is presented in

More information

RISK-SENSITIVE DYNAMIC PORTFOLIO OPTIMIZATION WITH PARTIAL INFORMATION ON INFINITE TIME HORIZON

RISK-SENSITIVE DYNAMIC PORTFOLIO OPTIMIZATION WITH PARTIAL INFORMATION ON INFINITE TIME HORIZON The Annals of Applied Probability 22, Vol. 12, No. 1, 173 195 RISK-SENSITIVE DYNAMIC PORTFOLIO OPTIMIZATION WITH PARTIAL INFORMATION ON INFINITE TIME HORIZON BY HIDEO NAGAI AND SHIGE PENG Osaka University

More information

A MAXIMUM PRINCIPLE FOR STOCHASTIC OPTIMAL CONTROL WITH TERMINAL STATE CONSTRAINTS, AND ITS APPLICATIONS

A MAXIMUM PRINCIPLE FOR STOCHASTIC OPTIMAL CONTROL WITH TERMINAL STATE CONSTRAINTS, AND ITS APPLICATIONS COMMUNICATIONS IN INFORMATION AND SYSTEMS c 2006 International Press Vol. 6, No. 4, pp. 321-338, 2006 004 A MAXIMUM PRINCIPLE FOR STOCHASTIC OPTIMAL CONTROL WITH TERMINAL STATE CONSTRAINTS, AND ITS APPLICATIONS

More information

On the Multi-Dimensional Controller and Stopper Games

On the Multi-Dimensional Controller and Stopper Games On the Multi-Dimensional Controller and Stopper Games Joint work with Yu-Jui Huang University of Michigan, Ann Arbor June 7, 2012 Outline Introduction 1 Introduction 2 3 4 5 Consider a zero-sum controller-and-stopper

More information

Information and Credit Risk

Information and Credit Risk Information and Credit Risk M. L. Bedini Université de Bretagne Occidentale, Brest - Friedrich Schiller Universität, Jena Jena, March 2011 M. L. Bedini (Université de Bretagne Occidentale, Brest Information

More information

I forgot to mention last time: in the Ito formula for two standard processes, putting

I forgot to mention last time: in the Ito formula for two standard processes, putting I forgot to mention last time: in the Ito formula for two standard processes, putting dx t = a t dt + b t db t dy t = α t dt + β t db t, and taking f(x, y = xy, one has f x = y, f y = x, and f xx = f yy

More information

A stochastic control approach to robust duality in utility maximization

A stochastic control approach to robust duality in utility maximization Dept. of Math./CMA University of Oslo Pure Mathematics No 5 ISSN 86 2439 April 213 A stochastic control approach to robust duality in utility maximization Bernt Øksendal 1,2 Agnès Sulem 3 18 April 213

More information

Weak Singular Optimal Controls Subject to State-Control Equality Constraints

Weak Singular Optimal Controls Subject to State-Control Equality Constraints Int. Journal of Math. Analysis, Vol. 6, 212, no. 36, 1797-1812 Weak Singular Optimal Controls Subject to State-Control Equality Constraints Javier F Rosenblueth IIMAS UNAM, Universidad Nacional Autónoma

More information

Stochastic optimal control with rough paths

Stochastic optimal control with rough paths Stochastic optimal control with rough paths Paul Gassiat TU Berlin Stochastic processes and their statistics in Finance, Okinawa, October 28, 2013 Joint work with Joscha Diehl and Peter Friz Introduction

More information

Control of McKean-Vlasov Dynamics versus Mean Field Games

Control of McKean-Vlasov Dynamics versus Mean Field Games Control of McKean-Vlasov Dynamics versus Mean Field Games René Carmona (a),, François Delarue (b), and Aimé Lachapelle (a), (a) ORFE, Bendheim Center for Finance, Princeton University, Princeton, NJ 8544,

More information

Prof. Erhan Bayraktar (University of Michigan)

Prof. Erhan Bayraktar (University of Michigan) September 17, 2012 KAP 414 2:15 PM- 3:15 PM Prof. (University of Michigan) Abstract: We consider a zero-sum stochastic differential controller-and-stopper game in which the state process is a controlled

More information

Optimal Trade Execution with Instantaneous Price Impact and Stochastic Resilience

Optimal Trade Execution with Instantaneous Price Impact and Stochastic Resilience Optimal Trade Execution with Instantaneous Price Impact and Stochastic Resilience Ulrich Horst 1 Humboldt-Universität zu Berlin Department of Mathematics and School of Business and Economics Vienna, Nov.

More information

Chapter Introduction. A. Bensoussan

Chapter Introduction. A. Bensoussan Chapter 2 EXPLICIT SOLUTIONS OFLINEAR QUADRATIC DIFFERENTIAL GAMES A. Bensoussan International Center for Decision and Risk Analysis University of Texas at Dallas School of Management, P.O. Box 830688

More information

Production and Relative Consumption

Production and Relative Consumption Mean Field Growth Modeling with Cobb-Douglas Production and Relative Consumption School of Mathematics and Statistics Carleton University Ottawa, Canada Mean Field Games and Related Topics III Henri Poincaré

More information

Linear Differential Equations. Problems

Linear Differential Equations. Problems Chapter 1 Linear Differential Equations. Problems 1.1 Introduction 1.1.1 Show that the function ϕ : R R, given by the expression ϕ(t) = 2e 3t for all t R, is a solution of the Initial Value Problem x =

More information

Solving Stochastic Partial Differential Equations as Stochastic Differential Equations in Infinite Dimensions - a Review

Solving Stochastic Partial Differential Equations as Stochastic Differential Equations in Infinite Dimensions - a Review Solving Stochastic Partial Differential Equations as Stochastic Differential Equations in Infinite Dimensions - a Review L. Gawarecki Kettering University NSF/CBMS Conference Analysis of Stochastic Partial

More information

Dynamic and Stochastic Brenier Transport via Hopf-Lax formulae on Was

Dynamic and Stochastic Brenier Transport via Hopf-Lax formulae on Was Dynamic and Stochastic Brenier Transport via Hopf-Lax formulae on Wasserstein Space With many discussions with Yann Brenier and Wilfrid Gangbo Brenierfest, IHP, January 9-13, 2017 ain points of the

More information

Question 1. The correct answers are: (a) (2) (b) (1) (c) (2) (d) (3) (e) (2) (f) (1) (g) (2) (h) (1)

Question 1. The correct answers are: (a) (2) (b) (1) (c) (2) (d) (3) (e) (2) (f) (1) (g) (2) (h) (1) Question 1 The correct answers are: a 2 b 1 c 2 d 3 e 2 f 1 g 2 h 1 Question 2 a Any probability measure Q equivalent to P on F 2 can be described by Q[{x 1, x 2 }] := q x1 q x1,x 2, 1 where q x1, q x1,x

More information

The concentration of a drug in blood. Exponential decay. Different realizations. Exponential decay with noise. dc(t) dt.

The concentration of a drug in blood. Exponential decay. Different realizations. Exponential decay with noise. dc(t) dt. The concentration of a drug in blood Exponential decay C12 concentration 2 4 6 8 1 C12 concentration 2 4 6 8 1 dc(t) dt = µc(t) C(t) = C()e µt 2 4 6 8 1 12 time in minutes 2 4 6 8 1 12 time in minutes

More information

Mathematical Economics. Lecture Notes (in extracts)

Mathematical Economics. Lecture Notes (in extracts) Prof. Dr. Frank Werner Faculty of Mathematics Institute of Mathematical Optimization (IMO) http://math.uni-magdeburg.de/ werner/math-ec-new.html Mathematical Economics Lecture Notes (in extracts) Winter

More information

Stochastic Maximum Principle and Dynamic Convex Duality in Continuous-time Constrained Portfolio Optimization

Stochastic Maximum Principle and Dynamic Convex Duality in Continuous-time Constrained Portfolio Optimization Stochastic Maximum Principle and Dynamic Convex Duality in Continuous-time Constrained Portfolio Optimization Yusong Li Department of Mathematics Imperial College London Thesis presented for examination

More information

Short-time expansions for close-to-the-money options under a Lévy jump model with stochastic volatility

Short-time expansions for close-to-the-money options under a Lévy jump model with stochastic volatility Short-time expansions for close-to-the-money options under a Lévy jump model with stochastic volatility José Enrique Figueroa-López 1 1 Department of Statistics Purdue University Statistics, Jump Processes,

More information

Mathematical Methods for Neurosciences. ENS - Master MVA Paris 6 - Master Maths-Bio ( )

Mathematical Methods for Neurosciences. ENS - Master MVA Paris 6 - Master Maths-Bio ( ) Mathematical Methods for Neurosciences. ENS - Master MVA Paris 6 - Master Maths-Bio (2014-2015) Etienne Tanré - Olivier Faugeras INRIA - Team Tosca November 26th, 2014 E. Tanré (INRIA - Team Tosca) Mathematical

More information

A Concise Course on Stochastic Partial Differential Equations

A Concise Course on Stochastic Partial Differential Equations A Concise Course on Stochastic Partial Differential Equations Michael Röckner Reference: C. Prevot, M. Röckner: Springer LN in Math. 1905, Berlin (2007) And see the references therein for the original

More information

Math Ordinary Differential Equations

Math Ordinary Differential Equations Math 411 - Ordinary Differential Equations Review Notes - 1 1 - Basic Theory A first order ordinary differential equation has the form x = f(t, x) (11) Here x = dx/dt Given an initial data x(t 0 ) = x

More information

Skorokhod Embeddings In Bounded Time

Skorokhod Embeddings In Bounded Time Skorokhod Embeddings In Bounded Time Stefan Ankirchner Institute for Applied Mathematics University of Bonn Endenicher Allee 6 D-535 Bonn ankirchner@hcm.uni-bonn.de Philipp Strack Bonn Graduate School

More information

Weak solutions of mean-field stochastic differential equations

Weak solutions of mean-field stochastic differential equations Weak solutions of mean-field stochastic differential equations Juan Li School of Mathematics and Statistics, Shandong University (Weihai), Weihai 26429, China. Email: juanli@sdu.edu.cn Based on joint works

More information

A class of globally solvable systems of BSDE and applications

A class of globally solvable systems of BSDE and applications A class of globally solvable systems of BSDE and applications Gordan Žitković Department of Mathematics The University of Texas at Austin Thera Stochastics - Wednesday, May 31st, 2017 joint work with Hao

More information

Hypothesis testing for Stochastic PDEs. Igor Cialenco

Hypothesis testing for Stochastic PDEs. Igor Cialenco Hypothesis testing for Stochastic PDEs Igor Cialenco Department of Applied Mathematics Illinois Institute of Technology igor@math.iit.edu Joint work with Liaosha Xu Research partially funded by NSF grants

More information

Quadratic BSDE systems and applications

Quadratic BSDE systems and applications Quadratic BSDE systems and applications Hao Xing London School of Economics joint work with Gordan Žitković Stochastic Analysis and Mathematical Finance-A Fruitful Partnership 25 May, 2016 1 / 23 Backward

More information

Optimisation and optimal control 1: Tutorial solutions (calculus of variations) t 3

Optimisation and optimal control 1: Tutorial solutions (calculus of variations) t 3 Optimisation and optimal control : Tutorial solutions calculus of variations) Question i) Here we get x d ) = d ) = 0 Hence ṫ = ct 3 where c is an arbitrary constant. Integrating, xt) = at 4 + b where

More information

New ideas in the non-equilibrium statistical physics and the micro approach to transportation flows

New ideas in the non-equilibrium statistical physics and the micro approach to transportation flows New ideas in the non-equilibrium statistical physics and the micro approach to transportation flows Plenary talk on the conference Stochastic and Analytic Methods in Mathematical Physics, Yerevan, Armenia,

More information

Stochastic Stackelberg equilibria with applications to time dependent newsvendor models

Stochastic Stackelberg equilibria with applications to time dependent newsvendor models INSTITUTT FOR FORTAKSØKONOMI DPARTMNT OF FINANC AND MANAGMNT SCINC FOR 9 2011 ISSN: 1500-4066 MAY 2011 Discussion paper Stochastic Stackelberg equilibria with applications to time dependent newsvendor

More information

The incompressible Navier-Stokes equations in vacuum

The incompressible Navier-Stokes equations in vacuum The incompressible, Université Paris-Est Créteil Piotr Bogus law Mucha, Warsaw University Journées Jeunes EDPistes 218, Institut Elie Cartan, Université de Lorraine March 23th, 218 Incompressible Navier-Stokes

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

Poisson Jumps in Credit Risk Modeling: a Partial Integro-differential Equation Formulation

Poisson Jumps in Credit Risk Modeling: a Partial Integro-differential Equation Formulation Poisson Jumps in Credit Risk Modeling: a Partial Integro-differential Equation Formulation Jingyi Zhu Department of Mathematics University of Utah zhu@math.utah.edu Collaborator: Marco Avellaneda (Courant

More information

Stochastic Differential Equations.

Stochastic Differential Equations. Chapter 3 Stochastic Differential Equations. 3.1 Existence and Uniqueness. One of the ways of constructing a Diffusion process is to solve the stochastic differential equation dx(t) = σ(t, x(t)) dβ(t)

More information

p 1 ( Y p dp) 1/p ( X p dp) 1 1 p

p 1 ( Y p dp) 1/p ( X p dp) 1 1 p Doob s inequality Let X(t) be a right continuous submartingale with respect to F(t), t 1 P(sup s t X(s) λ) 1 λ {sup s t X(s) λ} X + (t)dp 2 For 1 < p

More information

Continuous Time Finance

Continuous Time Finance Continuous Time Finance Lisbon 2013 Tomas Björk Stockholm School of Economics Tomas Björk, 2013 Contents Stochastic Calculus (Ch 4-5). Black-Scholes (Ch 6-7. Completeness and hedging (Ch 8-9. The martingale

More information

An Integral-type Constraint Qualification for Optimal Control Problems with State Constraints

An Integral-type Constraint Qualification for Optimal Control Problems with State Constraints An Integral-type Constraint Qualification for Optimal Control Problems with State Constraints S. Lopes, F. A. C. C. Fontes and M. d. R. de Pinho Officina Mathematica report, April 4, 27 Abstract Standard

More information

Continuous dependence estimates for the ergodic problem with an application to homogenization

Continuous dependence estimates for the ergodic problem with an application to homogenization Continuous dependence estimates for the ergodic problem with an application to homogenization Claudio Marchi Bayreuth, September 12 th, 2013 C. Marchi (Università di Padova) Continuous dependence Bayreuth,

More information

Dichotomy, the Closed Range Theorem and Optimal Control

Dichotomy, the Closed Range Theorem and Optimal Control Dichotomy, the Closed Range Theorem and Optimal Control Pavel Brunovský (joint work with Mária Holecyová) Comenius University Bratislava, Slovakia Praha 13. 5. 2016 Brunovsky Praha 13. 5. 2016 Closed Range

More information

1 Differential Equations

1 Differential Equations Reading [Simon], Chapter 24, p. 633-657. 1 Differential Equations 1.1 Definition and Examples A differential equation is an equation involving an unknown function (say y = y(t)) and one or more of its

More information

Equilibrium Conditions for the Simple New Keynesian Model

Equilibrium Conditions for the Simple New Keynesian Model Equilibrium Conditions for the Simple New Keynesian Model Lawrence J. Christiano August 4, 04 Baseline NK model with no capital and with a competitive labor market. private sector equilibrium conditions

More information

Bernt Øksendal, Leif Sandal and Jan Ubøe. 7 February Abstract

Bernt Øksendal, Leif Sandal and Jan Ubøe. 7 February Abstract Dept. of Math./CMA University of Oslo Pure Mathematics ISSN 0806 2439 February 2012 Stochastic Stackelberg equilibria with applications to time-dependent newsvendor models Bernt Øksendal, Leif Sandal and

More information

Uncertainty quantification and systemic risk

Uncertainty quantification and systemic risk Uncertainty quantification and systemic risk Josselin Garnier (Université Paris 7) with George Papanicolaou and Tzu-Wei Yang (Stanford University) April 9, 2013 Modeling Systemic Risk We consider evolving

More information

Theoretical Tutorial Session 2

Theoretical Tutorial Session 2 1 / 36 Theoretical Tutorial Session 2 Xiaoming Song Department of Mathematics Drexel University July 27, 216 Outline 2 / 36 Itô s formula Martingale representation theorem Stochastic differential equations

More information

Bernardo D Auria Stochastic Processes /12. Notes. March 29 th, 2012

Bernardo D Auria Stochastic Processes /12. Notes. March 29 th, 2012 1 Stochastic Calculus Notes March 9 th, 1 In 19, Bachelier proposed for the Paris stock exchange a model for the fluctuations affecting the price X(t) of an asset that was given by the Brownian motion.

More information

Maximum principle for a stochastic delayed system involving terminal state constraints

Maximum principle for a stochastic delayed system involving terminal state constraints Wen and Shi Journal of Inequalities and Applications 217) 217:13 DOI 1.1186/s1366-17-1378-z R E S E A R C H Open Access Maximum principle for a stochastic delayed system involving terminal state constraints

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

Errata for Stochastic Calculus for Finance II Continuous-Time Models September 2006

Errata for Stochastic Calculus for Finance II Continuous-Time Models September 2006 1 Errata for Stochastic Calculus for Finance II Continuous-Time Models September 6 Page 6, lines 1, 3 and 7 from bottom. eplace A n,m by S n,m. Page 1, line 1. After Borel measurable. insert the sentence

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

Hybrid Constrained Estimation For Linear Time-Varying Systems

Hybrid Constrained Estimation For Linear Time-Varying Systems 2018 IEEE Conference on Decision and Control (CDC) Miami Beach, FL, USA, Dec. 17-19, 2018 Hybrid Constrained Estimation For Linear Time-Varying Systems Soulaimane Berkane, Abdelhamid Tayebi and Andrew

More information

Man Kyu Im*, Un Cig Ji **, and Jae Hee Kim ***

Man Kyu Im*, Un Cig Ji **, and Jae Hee Kim *** JOURNAL OF THE CHUNGCHEONG MATHEMATICAL SOCIETY Volume 19, No. 4, December 26 GIRSANOV THEOREM FOR GAUSSIAN PROCESS WITH INDEPENDENT INCREMENTS Man Kyu Im*, Un Cig Ji **, and Jae Hee Kim *** Abstract.

More information

Convergence of Particle Filtering Method for Nonlinear Estimation of Vortex Dynamics

Convergence of Particle Filtering Method for Nonlinear Estimation of Vortex Dynamics Convergence of Particle Filtering Method for Nonlinear Estimation of Vortex Dynamics Meng Xu Department of Mathematics University of Wyoming February 20, 2010 Outline 1 Nonlinear Filtering Stochastic Vortex

More information

L2 gains and system approximation quality 1

L2 gains and system approximation quality 1 Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science 6.242, Fall 24: MODEL REDUCTION L2 gains and system approximation quality 1 This lecture discusses the utility

More information

Sufficiency for Essentially Bounded Controls Which Do Not Satisfy the Strengthened Legendre-Clebsch Condition

Sufficiency for Essentially Bounded Controls Which Do Not Satisfy the Strengthened Legendre-Clebsch Condition Applied Mathematical Sciences, Vol. 12, 218, no. 27, 1297-1315 HIKARI Ltd, www.m-hikari.com https://doi.org/1.12988/ams.218.81137 Sufficiency for Essentially Bounded Controls Which Do Not Satisfy the Strengthened

More information

Optimal Control Theory

Optimal Control Theory Adv. Scientific Computing III Optimal Control Theory Catalin Trenchea Department of Mathematics University of Pittsburgh Fall 2008 Adv. Scientific Computing III > Control systems in Banach spaces Let X

More information

On Irregular Linear Quadratic Control: Deterministic Case

On Irregular Linear Quadratic Control: Deterministic Case 1 On Irregular Linear Quadratic Control: Deterministic Case Huanshui Zhang and Juanjuan Xu arxiv:1711.9213v2 math.oc 1 Mar 218 Abstract This paper is concerned with fundamental optimal linear quadratic

More information

Mean-Field optimization problems and non-anticipative optimal transport. Beatrice Acciaio

Mean-Field optimization problems and non-anticipative optimal transport. Beatrice Acciaio Mean-Field optimization problems and non-anticipative optimal transport Beatrice Acciaio London School of Economics based on ongoing projects with J. Backhoff, R. Carmona and P. Wang Robust Methods in

More information

UNCERTAINTY FUNCTIONAL DIFFERENTIAL EQUATIONS FOR FINANCE

UNCERTAINTY FUNCTIONAL DIFFERENTIAL EQUATIONS FOR FINANCE Surveys in Mathematics and its Applications ISSN 1842-6298 (electronic), 1843-7265 (print) Volume 5 (2010), 275 284 UNCERTAINTY FUNCTIONAL DIFFERENTIAL EQUATIONS FOR FINANCE Iuliana Carmen Bărbăcioru Abstract.

More information

A TWO PARAMETERS AMBROSETTI PRODI PROBLEM*

A TWO PARAMETERS AMBROSETTI PRODI PROBLEM* PORTUGALIAE MATHEMATICA Vol. 53 Fasc. 3 1996 A TWO PARAMETERS AMBROSETTI PRODI PROBLEM* C. De Coster** and P. Habets 1 Introduction The study of the Ambrosetti Prodi problem has started with the paper

More information

1 Brownian Local Time

1 Brownian Local Time 1 Brownian Local Time We first begin by defining the space and variables for Brownian local time. Let W t be a standard 1-D Wiener process. We know that for the set, {t : W t = } P (µ{t : W t = } = ) =

More information

MATH 220 solution to homework 5

MATH 220 solution to homework 5 MATH 220 solution to homework 5 Problem. (i Define E(t = k(t + p(t = then E (t = 2 = 2 = 2 u t u tt + u x u xt dx u 2 t + u 2 x dx, u t u xx + u x u xt dx x [u tu x ] dx. Because f and g are compactly

More information

Problem 1. Construct a filtered probability space on which a Brownian motion W and an adapted process X are defined and such that

Problem 1. Construct a filtered probability space on which a Brownian motion W and an adapted process X are defined and such that Stochatic Calculu Example heet 4 - Lent 5 Michael Tehranchi Problem. Contruct a filtered probability pace on which a Brownian motion W and an adapted proce X are defined and uch that dx t = X t t dt +

More information

Numerical solution of second-order stochastic differential equations with Gaussian random parameters

Numerical solution of second-order stochastic differential equations with Gaussian random parameters Journal of Linear and Topological Algebra Vol. 2, No. 4, 23, 223-235 Numerical solution of second-order stochastic differential equations with Gaussian random parameters R. Farnoosh a, H. Rezazadeh a,,

More information

Collusion between the government and the enterprise in pollution management

Collusion between the government and the enterprise in pollution management 30 5 015 10 JOURNAL OF SYSTEMS ENGINEERING Vol.30 No.5 Oct. 015 1, (1., 400044;., 400044 :. Stackelberg,.,,,. : ; Stackelberg ; ; ; : F4 : A : 1000 5781(01505 0584 10 doi: 10.13383/j.cnki.jse.015.05.00

More information

Observer design for a general class of triangular systems

Observer design for a general class of triangular systems 1st International Symposium on Mathematical Theory of Networks and Systems July 7-11, 014. Observer design for a general class of triangular systems Dimitris Boskos 1 John Tsinias Abstract The paper deals

More information

Schrodinger Bridges classical and quantum evolution

Schrodinger Bridges classical and quantum evolution .. Schrodinger Bridges classical and quantum evolution Tryphon Georgiou University of Minnesota! Joint work with Michele Pavon and Yongxin Chen! LCCC Workshop on Dynamic and Control in Networks Lund, October

More information

Entrance Exam, Differential Equations April, (Solve exactly 6 out of the 8 problems) y + 2y + y cos(x 2 y) = 0, y(0) = 2, y (0) = 4.

Entrance Exam, Differential Equations April, (Solve exactly 6 out of the 8 problems) y + 2y + y cos(x 2 y) = 0, y(0) = 2, y (0) = 4. Entrance Exam, Differential Equations April, 7 (Solve exactly 6 out of the 8 problems). Consider the following initial value problem: { y + y + y cos(x y) =, y() = y. Find all the values y such that the

More information

Stochastic contraction BACS Workshop Chamonix, January 14, 2008

Stochastic contraction BACS Workshop Chamonix, January 14, 2008 Stochastic contraction BACS Workshop Chamonix, January 14, 2008 Q.-C. Pham N. Tabareau J.-J. Slotine Q.-C. Pham, N. Tabareau, J.-J. Slotine () Stochastic contraction 1 / 19 Why stochastic contraction?

More information

Exercises. T 2T. e ita φ(t)dt.

Exercises. T 2T. e ita φ(t)dt. Exercises. Set #. Construct an example of a sequence of probability measures P n on R which converge weakly to a probability measure P but so that the first moments m,n = xdp n do not converge to m = xdp.

More information

A Lieb-Robinson Bound for the Toda System

A Lieb-Robinson Bound for the Toda System 1 FRG Workshop May 18, 2011 A Lieb-Robinson Bound for the Toda System Umar Islambekov The University of Arizona joint work with Robert Sims 2 Outline: 1. Motivation 2. Toda Lattice 3. Lieb-Robinson bounds

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

Bernardo D Auria Stochastic Processes /10. Notes. Abril 13 th, 2010

Bernardo D Auria Stochastic Processes /10. Notes. Abril 13 th, 2010 1 Stochastic Calculus Notes Abril 13 th, 1 As we have seen in previous lessons, the stochastic integral with respect to the Brownian motion shows a behavior different from the classical Riemann-Stieltjes

More information

Optimal transportation and optimal control in a finite horizon framework

Optimal transportation and optimal control in a finite horizon framework Optimal transportation and optimal control in a finite horizon framework Guillaume Carlier and Aimé Lachapelle Université Paris-Dauphine, CEREMADE July 2008 1 MOTIVATIONS - A commitment problem (1) Micro

More information

Theory of Ordinary Differential Equations

Theory of Ordinary Differential Equations Theory of Ordinary Differential Equations Existence, Uniqueness and Stability Jishan Hu and Wei-Ping Li Department of Mathematics The Hong Kong University of Science and Technology ii Copyright c 24 by

More information