SUBORDINATED PROCESSES AND CAUCHY PROBLEMS by ERKAN NANE DEPARTMENT OF STATISTICS AND PROBABILITY MICHIGAN STATE UNIVERSITY OUTLINE • Introduction and history • Brownian subordinators • Cauchy problems on bounded domains • Other subordinators • Scaling Limits • Conclusion and open problems 2 INTRODUCTION AND HISTORY In recent years, starting with the articles of Burdzy (1993) and (1994), researchers had interest in iterated processes in which one changes the time parameter with one-dimensional Brownian motion. To define iterated Brownian motion Zt, due to Burdzy (1993), started at z ∈ R, let Xt+, Xt− and Yt be three independent one-dimensional Brownian motions, all started at 0. Two-sided Brownian motion is defined to be Xt = Xt+, t ≥ 0 − X(−t) , t < 0. Then iterated Brownian motion started at z ∈ R is Zt = z + X(Yt), t ≥ 0. 3 BM versus IBM: This process has many properties analogous to those of Brownian motion; we list a few (1) Zt has stationary (but not independent) increments, and is a self-similar process of index 1/4. (2) Laws of the iterated logarithm (LIL) holds: usual LIL by Burdzy (1993) 25/4 Z(t) lim sup 1/4 = 3/4 a.s. 3/4 3 t→∞ t (log log(1/t)) Chung-type LIL by Khoshnevisan and Lewis (1996) and Hu et al. (1995). (3) Khoshnevisan and Lewis (1999) extended results of Burdzy (1994), to develop a stochastic calculus for iterated Brownian motion. 4 (4) In 1998, Burdzy and Khoshnevisan showed that IBM can be used to model diffusion in a crack. (5) Local time of this process was studied by Burdzy and Khoshnevisan (1995), Csáki, Csörgö, Földes, and Révész (1996), Shi and Yor (1997), Xiao (1998), and Hu (1999). (6) Bañuelos and DeBlassie (2006) studied the distribution of exit place for iterated Brownian motion in cones. (7) DeBlassie (2004) studied the lifetime asymptotics of iterated Brownian motion in cones and Bounded domains. Nane (2006), in a series of papers, extended some of the results of DeBlassie. He also studied the lifetime asymtotics of iterated Brownian motion in several unbounded domains(parabolashaped domains, twisted domains...). 5 (8) Khoshnevisan and Lewis (1996) established the modulus of continuity for iterated Brownian motion: with probability one |Z(s) − Z(t)| lim sup sup = 1. 1/4 3/4 δ→0 0≤s,t≤1 0≤|s−t|≤δ δ (log(1/δ)) 6 0.2 0.15 0.1 0.05 0 0.05 0.1 0.15 0.2 0.25 t –0.05 –0.1 7 Figure 1: Simulations of two Brownian motions 0.3 0.1 0.05 0 0.05 0.1 0.15 0.2 0.25 t –0.05 8 Figure 2: Simulation of IBM Zt1 = X(|Yt|) 0.3 PDE connection The classical well-known connection of a PDE and a stochastic process is the Brownian motion and heat equation connection. Let Xt ∈ Rd be Brownian motion started at x. Then the function u(t, x) = Ex[f (Xt)] solves the Cauchy problem ∂ u(t, x) = ∆u(t, x), t > 0, x ∈ Rd ∂t u(0, x) = f (x), x ∈ Rd. In addition to the above properties of IBM there is an interesting connection between iterated Brownian motion and the biharmonic operator ∆2; the function 9 u(t, x) = Ex[f (Zt)] solves the Cauchy problem (Allouba and Zheng (2001) and DeBlassie (2004)) ∂ ∆f (x) u(t, x) = √ + ∆2u(t, x); ∂t πt u(0, x) = f (x). (1) for t > 0 and x ∈ Rd. The non-Markovian property of IBM is reflected by the appearance of the initial function f (x) in the PDE. 2 Let q(t, s) = √ 2 exp − s4t be the transition density of 4πt reflected one-dimensional Brownian motion, |Bt|. The essential argument, using integration by parts twice, is that 10 ∂ ∂t u(t, x) ∞ ∂ = T (s)f (x) q(t, s)ds ∂t Z0 ∞ ∂2 = T (s)f (x) 2 q(t, s)ds ∂s 0 ∂ = q(t, s) [T (s)f (x)] ∂s s=0 Z ∞ 2 ∂ + [T (s)f (x)] q(t, s)ds 2 ∂s 0 Z Z ∞ = q(t, 0)Lx[T (0)f (x)] + L2x [T (s)f (x)] q(t, s)ds Z ∞ 0 1 = √ Lxf (x) + L2x T (s)f (x) q(t, s)ds πt 0 11 Fractional Diffusion Nigmatullin (1986) gave a Physical derivation of fractional diffusion ∂β u(t, x) = Lxu(t, x); β ∂t u(0, x) = f (x) (2) where 0 < β < 1 and Lx is the generator of some continuous Markov process X0(t) started at x = 0. Here ∂ β g(t)/∂tβ is the Caputo fractional derivative in time, which can be defined β β−1 as the inverse Laplace transform of s g̃(s) − s g(0), with R ∞ −st g̃(s) = 0 e g(t)dt the usual Laplace transform. Zaslavsky (1994) used this to model Hamiltonian chaos. 12 Stochastic solution Baeumer and Meerschaert (2001) and Meerschaert and Scheffler (2004) shows that, in the case p(t, x) = T (t)f (x) is a bounded continuous semigroup on a Banach space (with corresponding process Xt, Et = inf{u : Du > t}, Dt is a stable subordinator with index β ) , the formula t u(t, x) = Ex(f (XEt )) = β Z ∞ 0 p(s, x)gβ ( t −1/β−1 )s ds 1/β s yields a solution to the fractional Cauchy problem: ∂β u(t, x) = Lxu(t, x); u(0, x) = f (x) (3) β ∂t Here gβ (t) is the smooth density of the stable subordinator, such R ∞ −st β that the Laplace transform g̃β (s) = 0 e gβ (t) dt = e−s . 13 BROWNIAN SUBORDINATORS (Baeumer, Meerschaert and Nane (2007)) show that, taking β = 1/2 in the time-fractional diffusion yields exactly 1-D d distributions X|Bt| = XEt . If Lx is the generator of the semigroup T (t)f (x) = Ex[(f (Xt))] on L1(Rd), then for any f ∈ D(Lx) ∂ Lxf (x) + Lx2u(t, x); u(t, x) = √ ∂t πt u(0, x) = f (x), (4) and the fractional Cauchy problem (3) with β = 1/2 have the same solution 2 u(t, x) = Ex[f (X(|Bt|))] = √ 4πt 14 Z ∞ 0 2 s T (s)f (x) exp − ds. 4t Fourier-Laplace method The Lévy process X0(t) has characteristic function with E[exp(ik · X0(t))] = exp(tψ(k)) 1 ψ(k) = ik·a− k·Qk+ 2 Z y6=0 eik·y − 1 − ik · y ν(dy), 2 1 + ||y|| where a ∈ Rd, Q is a nonnegative definite matrix, and ν is a d σ -finite Borel measure on R such that Z y6=0 min{1, ||y||2}ν(dy) < ∞. Denote the Fourier transform by fˆ(k) = Z e−ik·xf (x) dx Rd 15 Meerschaert and Scheffler (2001) shows that Lxf (x) is the inverse Fourier transform of ψ(k)fˆ(k) for all f ∈ D(Lx), where D(Lx) = {f ∈ L1(Rd) : ψ(k)fˆ(k) = ĥ(k) ∃ h ∈ L1(Rd)}, and 1 Lxf (x) = a · ∇f (x) + ∇ · Q∇f (x) 2 Z ∇f (x) · y + f (x + y) − f (x) − ν(dy) 2 1+y y6=0 (5) for all f ∈ W 2,1(Rd), the Sobolev space of L1-functions whose first and second partial derivatives are all L1-functions. We can also write Lx (∂/∂x1, . . . , ∂/∂xd)′. = 16 ψ(−i∇) where ∇ = For example, if X0(t) is spherically symmetric stable then ψ(k) = −Dkkkα and Lx = −D(−∆)α/2, a fractional derivative in space, using the correspondence kj → −i∂/∂xj for 1 ≤ j ≤ d. If X0 has independent marginals, P then one possible P stable form is ψ(k) = D j (ikj )αj and Lx = D j ∂ αj /∂xα j using Riemann-Liouville fractional derivatives in each variable. This form does not coincide with the fractional Laplacian unless all α j = 2. 17 The proof does not use Theorem 0.1 in Allouba and Zheng (2001), rather it relies on a Laplace-Fourier transform argument. We will use the following notation for the Laplace, Fourier, and Fourier-Laplace transforms (respectively): ũ(s, x) = Z ∞ e−stu(t, x)dt; Z0 e−ik·xu(t, x)dx; ZR d Z ∞ e−ik·x e−stu(t, x)dtdx. ū(s, k) = û(t, k) = Rd 0 18 Let ψ be the characteristic exponent of Xt. transforms on both sides of (4) to get Take Fourier ∂ û(t, k) 1 = √ ψ(k)fˆ(k) + ψ(k)2û(t, k) ∂t πt using the fact that ψ(k)fˆ(k) is the Fourier transform of Lxf (x). Then take Laplace transforms on both sides to get sū(s, k) − û(t = 0, k) = s−1/2ψ(k)fˆ(k) + ψ(k)2ū(s, k), using the well-known Laplace transform formula Z ∞ 0 t−β e−stdt = sβ−1, β < 1. Γ(1 − β) 19 Since û(t = 0, k) = fˆ(k), collecting like terms yields (1 + s−1/2ψ(k))fˆ(k) ū(s, k) = s − ψ(k)2 (6) for s > 0 sufficiently large. On the other hand, taking Fourier transforms on both sides of (3) with β = 1/2 gives ∂ 1/2û(t, k) = ψ(k)û(t, k) 1/2 ∂t Take Laplace transforms on both sides, using the fact that sβ g̃(s) − sβ−1g(0) is the Laplace transform of the Caputo fractional derivative ∂ β g(t)/∂tβ , to get s1/2ū(s, k) − s−1/2fˆ(k) = ψ(k)ū(s, k) 20 and collect terms to obtain s−1/2fˆ(k) ū(s, k) = 1/2 s − ψ(k) s−1/2fˆ(k) s1/2 + ψ(k) = 1/2 · 1/2 s − ψ(k) s + ψ(k) (1 + s−1/2ψ(k))fˆ(k) = s − ψ(k)2 (7) which agrees with (6). For any fixed k ∈ Rd, the two formulae are well-defined and equal for all s > 0 sufficiently large. An easy extension of the argument as above shows that, under the same conditions, for any k = 2, 3, 4, . . . both the Cauchy 21 problem k−1 1−j/k X ∂u(t, x) t = Ljxf (x) + Lkxu(t, x); ∂t Γ(j/k) j=1 (8) u(0, x) = f (x) and the fractional Cauchy problem: ∂ 1/k u(t, x) = Lxu(t, x); 1/k ∂t u(0, x) = f (x), (9) have the same unique solution given by u(t, x) = Z ∞ p((t/s)β , x)gβ (s) ds 0 with β = 1/k . Hence the process Zt = X(Et) is also the stochastic solution to this higher order Cauchy problem. 22 Orsingher and Benghin (2004) and (2008) show that for β = 1/2n the solution to 1/2n ∂ n u(t, x) = ∆x u(t, x); 1/2 ∂t u(0, x) = f (x), (10) is given by running In(t) = B1(|B2(|B3(| · · · (Bn+1(t)) · · · |)|)|) Where Bj ’s are independent Brownian motions, i.e., u(t, x) = Ex(f (In(t))) solves (10), and solves (8) for k = 2n. 23 CAUCHY PROBLEMS ON BOUNDED DOMAINS Since we are working on a bounded domain, the Fourier transform methods are not useful. Instead we will employ Hilbert space methods. Hence, given a complete orthonormal basis {ψn(x)} on L2(D), we will call ū(t, n) = Z ψn(x)u(t, x)dx; ZD Z ∞ Z û(s, n) = ψn(x) e−stu(t, x)dtdx = ψn(x)ũ(s, x)dx. D D 0 the ψn, and ψn-Laplace transforms, respectively. Let D be bounded and every point of ∂D be regular for D C . The corresponding Markov process is a killed Brownian motion. We denote the eigenvalues and the eigenfunctions of ∆D by 24 ∞ {λn, φn}∞ n=1, where φn ∈ C (D). The corresponding heat kernel is given by pD (t, x, y) = ∞ X e−λntφn(x)φn(y). n=1 The series converges absolutely and uniformly on [t0, ∞) × D × D for all t0 > 0. In this case, the semigroup given by TD (t)f (x) = Ex[f (Xt)I(t < τD (X))] = = ∞ X Z pD (t, x, y)f (y)dy D e−λntφn(x)f¯(n) n=1 solves the Heat equation in D with Dirichlet boundary 25 conditions: ∂u(t, x) = ∆u(t, x), x ∈ D, t > 0, ∂t u(t, x) = 0, x ∈ ∂D, u(0, x) = f (x), x ∈ D. 26 Fractional Cauchy problems in bounded domains Let β ∈ (0, 1), D∞ = (0, ∞) × D and define ∂ ∂β H∆(D∞) ≡ u : D∞ → R : u, β u, ∆u ∈ C(D∞), ∂t ∂t ∂ u(t, x) ≤ g(x)tβ−1, g ∈ L∞(D), t > 0 . ∂t Let 0 < γ < 1. Let D be a bounded domain with ∂D ∈ C 1,γ , and TD (t) be the killed semigroup of Brownian motion {Xt} in D . Let Et be the process inverse to a stable subordinator of index β ∈ (0, 1) independent of {Xt}. Let f ∈ D(∆D ) ∩ C 1(D̄) ∩ C 2(D) for which the eigenfunction expansion (of ∆f ) with respect to the complete orthonormal basis {φn : n ∈ N} 27 converges uniformly and absolutely. Then the unique (classical) solution of u ∈ H∆(D∞) ∩ Cb(D̄∞) ∩ C 1(D̄) ∂β u(t, x) = ∆u(t, x); x ∈ D, t > 0 β ∂t u(t, x) = 0, x ∈ ∂D, t > 0, u(0, x) = f (x), x ∈ D. 28 (11) is given by u(t, x) = ∞ X f¯(n)φn(x)Eβ (−λntβ ) n=1 = Ex[f (X(Et))I(τD (X) > Et)] = ExZ[f (X(Et))I(τD (X(E)) > t)] t ∞ TD (l)f (x)gβ (tl−1/β )l−1/β−1dl = β 0 Z ∞ = TD ((t/l)β )f (x)gβ (l)dl. 0 Joint work with Meerschaert and Vellaisamy (2008). Analytic solution in intervals (0, M ) ⊂ R was obtained by Agrawal (2002). 29 Assume that u(t, x) solves (11). Taking φn- transforms in (11) we obtain ∂β ū(t, n) = −λnū(t, n). β ∂t (12) taking Laplace transforms on both sides of (12), we get sβ û(s, n) − sβ−1ū(0, n) = −λnû(s, n) (13) f¯(n)sβ−1 û(s, n) = β . s + λn (14) which leads to By inverting the above Laplace transform, we obtain ū(t, n) = f¯(n)Eβ (−λntβ ) 30 in terms of the Mittag-Leffler function defined by Eβ (z) = ∞ X k=0 zk . Γ(1 + βk) Inverting now the φn-transform, we get an L2-convergent solution of Equation (11) as (for each t ≥ 0 ) u(t, x) = ∞ X f¯(n)φn(x)Eβ (−λntβ ) n=1 31 (15) IBM in bounded domains Let Zt = X(|Yt|) be the iterated Brownian motion, D∞ = (0, ∞) × D and define ∂ u, ∆2u ∈ C(D∞), ∆u ∈ C 1(D̄) , H∆2 (D∞) ≡ u : D∞ → R : ∂t ∂ u(t, x) ≤ g(x)t−1/2, g ∈ L∞(D), t > 0 . ∂t Let D be a domain with ∂D ∈ C 1,γ , 0 < γ < 1. Let {Xt} be Brownian motion in Rd, and {Yt} be an independent Brownian motion in R. Let {Et} be the process inverse to a stable subordinator of index β = 1/2 independent of {Xt}. Let f ∈ D(∆D ) ∩ C 1(D̄) ∩ C 2(D)(⊂ L2(D)) be 32 such that the eigenfunction expansion of ∆f with respect to {φn : n ≥ 1} converges absolutely and uniformly. Then the (classical) solution of u ∈ H∆2 (D∞) ∩ Cb(D̄∞) ∩ C 1(D̄); ∂ ∆f (x) u(t, x) = √ + ∆2u(t, x), x ∈ D, t > 0; (16) ∂t πt u(t, x) = ∆u(t, x) = 0, t ≥ 0, x ∈ ∂D; u(0, x) = f (x), x ∈ D 33 is given by u(t, x) = Ex[f (Zt)I(τD (X) > |Yt|)] = Ex[f (X(Et))I(τD (X) > Et)] = EZx[f (X(Et))I(τD (X(E)) > t)] ∞ =2 TD (l)f (x)p(t, l)dl, (17) 0 where TD (l) is the heat semigroup in D, and p(t, l) is the transition density of one-dimensional Brownian motion {Yt}. Proof uses again equivalence with fractional Cauchy problem for β = 1/2. 34 OTHER SUBORDINATORS A Lévy process S = {St, t ≥ 0} with values in R is called strictly stable of index α ∈ (0, 2] if its characteristic function is given by πα α 1 + iνsgn(ξ) tan( 2 ) E exp(iξSt) = exp −t|ξ| , (18) χ where −1 ≤ ν ≤ 1 and χ > 0 are constants. When α = 2 and χ = 2, S is Brownian motion. 35 For any Borel set I ⊆ R, the occupation measure of S on I is defined by µI (A) = λ1{t ∈ I : St ∈ A} (19) for all Borel sets A ⊆ R, where λ1 is the one-dimensional Lebesgue measure. If µI is absolutely continuous with respect to the Lebesgue measure λ1 on R, we say that S has a local time on I and define its local time L(x, I) to be the Radon-Nikodým derivative of µI with respect to λ1, i.e., dµI L(x, I) = (x), dλ1 ∀x ∈ R. In the above, x is the so-called space variable, and I is the time variable of the local time. If I = [0, t], we will write L(x, I) as L(x, t). Moreover, if x = 0 then we will simply write L(0, t) as Lt . 36 Local time as a subordinator It is well-known (see, e.g. Bertoin (1996)) that a strictly stable Lévy process S has a local time if and only if α ∈ (1, 2]. It is well-known (see, e.g. Bertoin (1996)) that the inverse of a local time Lt of S is a stable subordinator: Gt = inf{u : Lu > t}, then Gt = ρDt, where Dt is a stable subordinator of index β = 1 − 1/α. ρ = π −1Γ(1+1/α)Γ(1−1/α)χ1/αRe{[1+iν tan(πα/2)]−1/α}. Hence Lt = Et/ρ where β = 1 − 1/α for some c > 0. For β = 1 − 1/α, c > 0, u(t, x) = Ex[f (X(Lt))] solves ∂β u(t, x) = cLxu(t, x); β ∂t 37 u(0, x) = f (x) (20) Symmetric stable subordinators α-time process is a Markov process subordinated to the absolute value of an independent one-dimensional symmetric α-stable process: Zt = X(|St|), where Xt is a Markov process and St is an independent symmetric α-stable process both started at 0. This process is self similar with index 1/2α when the outer process X is a Brownian motion. In this case Nane (2006) defined the Local time of this process and obtained Laws of the iterated logarithm for the local time for large time. 38 PDE-connection: Theorem 3 [Nane (2008)] Let T (s)f (x) = Ex[f (X(s))] be the semigroup of the continuous Markov process X(t) and let Lx be its generator. Let α = 1. Let f be a bounded measurable function in the domain of Lx, with Dij f bounded and Hölder continuous for all 1 ≤ i, j ≤ n. Then u(t, x) = Ex[f (Zt)] solves ∂2 2Lxf (x) 2 u(t, x) = − − L xu(t, x); 2 ∂t πt u(0, x) = f (x). 39 For α = l/m 6= 1 rational: the PDE is more complicated since kernels of symmetric α-stable processes satisfy a higher order PDE: 2m ∂ ∂ 2l l+1 α + (−1) p t (0, s) = 0. 2l 2m ∂s ∂t We also have to assume that we can integrate under the integral as much as we need in the case where the outer process is BM (or in general we can take the operator out of the integral). This is valid for α = 1/m, m = 2, 3, · · · by a Lemma in Nane (2008). 40 Theorem 4 [Nane (2008)] Let α ∈ (0, 2) be rational α = l/m, where l and m are relatively prime. Let T (s)f (x) = Ex[f (X(s))] be the semigroup of the continuous Markov process X(t) and let Lx be its generator. Let f be a bounded measurable function in the domain of Lx, with D γ f bounded and Hölder continuous for all multi index γ such that |γ| = 2l. Then u(t, x) = Ex[f (Zt)] solves l+1 (−1) 2m ∂ u(t, x) = −2 2m ∂t l 2l−2i X ∂ α 2i−1 p (0, s)| L f (x) s=0 t x 2l−2i ∂s i=1 L2l x u(t, x); − u(0, x) = f (x). 41 SCALING LIMITS If Sn = X1 + X2 + · · · + Xn is particle location at time n then the scaling limit is r −1/αS[rt] =⇒ Xt. The limit process Xt is called an α-stable Lévy motion. Another random walk Jn = T1 + T2 + · · · + Tn records the jump times. If P [Tn > t] ≈ t−β for 0 < β < 1 then r −1/β J[rt] =⇒ Dt. Dt is an increasing β -stable Lévy motion(a stable subordinator). Subordinator can give random time change: X(t) → X(Dt). 42 Inverse Stable subordinator Waiting time random walk has scaling limit r −1/β J[rt] =⇒ Dt. The number of jumps by time t is Nt = max{n > 0 : Jn ≤ t}. Renewal process inverse to random walk {Nt ≥ n} = {Jn ≤ t} Inverse process has inverse scaling limit r −β N[rt] =⇒ Et. Inverse stable process {Et ≤ u} = {Du ≥ t} yields the density formula: d d p(u, t) = P [Et ≤ u] = P [Du ≥ t] du du 43 CTRW scaling limits Particle jump random walk has scaling limit r −1/αS[rt] =⇒ Xt. Number of jumps has scaling limit r −β N[rt] =⇒ Et. CTRW is a random walk subordinated to a renewal process SNt = X1 + X2 + · · · + XNt CTRW scaling limit is a subordinated process: r−β/αSNct = (rβ )−1/αSrβ ·r−β Nct ≈ (rβ )−1/αSrβ Et =⇒ XEt CTRW scaling limit is not Markov, increments are not stationary. Meerschaert and collaborators (2004). 44 OPEN PROBLEMS Question 1. For β = 1/2, the inverse stable subordinator process Et and the process |Yt|, where Yt is a one-dimensional Brownian motion have the same transition density. Is there a similar correspondence between Et for β 6= 1/2 and other symmetric α-stable process Yt for 1 < α < 2. Question 2. Looking at the governing PDE for subordinators other than Brownian motion, are there any fractional in time PDE which has the same solution as the higher order pde? Question 3. Are there PDE connections of the iterated processes in bounded domain as the PDE connection of Brownian motion in bounded domains? 45 THANK YOU 46 REFERENCES O. P. Agrawal, Solution for a fractional diffusion-wave equation defined in a bounded domain. Fractional order calculus and its applications. Nonlinear Dynam. 29 (2002), no. 1-4, 145–155. H. Allouba and W. Zheng,Brownian-time processes: The pde connection and the half-derivative generator, Ann. Prob. 29 (2001), no. 2, 1780-1795. B. Baeumer, M. M. Meerschaert and E. Nane, Brownian subordinators and fractional Cauchy problems, Trans. Amer. Math. Soc. (to appear). 47 B. Baeumer and M.M. Meerschaert, Stochastic solutions for fractional Cauchy problems, Fractional Calculus Appl. Anal. 4 (2001), 481-500. C. Bandle, Isoperimetric Inequalities and Applications, Monog. Stud. Math. 7, Pitnam, Boston, 1980. P. Becker-Kern, M. M. Meerschaert and H. P. Scheffler, Limit theorems for coupled continuous time random walks, Ann. Probab. 32 (2004), 730–756. K. Burdzy, Some path properties of iterated Brownian motion, In: Seminar on Stochastic Processes (E. Çinlar, K.L. Chung and M.J. Sharpe, eds.), pp. 67–87, Birkhäuser, Boston, 1993. K. Burdzy, Variation of iterated Brownian motion, In: Workshops and Conference on Measure-valued Processes, Stochastic 48 Partial Differential Equations and Interacting Particle Systems (D.A. Dawson, ed.), pp. 35–53, Amer. Math. Soc. Providence, RI, 1994. K. Burdzy and D. Khoshnevisan, The level set of iterated Brownian motion, Séminarie de probabilités XXIX (Eds.: J Azéma, M. Emery, P.-A. Meyer and M. Yor), pp. 231-236, Lecture Notes in Mathematics, 1613, Springer, Berlin, 1995. K. Burdzy and D. Khoshnevisan, Brownian motion in a Brownian crack, Ann. Appl. Probab. 8 (1998), 708–748. E. Csáki, A. Földes and P. Révész, Strassen theorems for a class of iterated processes, Trans. Amer. Math. Soc. 349 (1997), 1153–1167. 49 E. Csáki, M. Csörgö, A. Földes, and P. Révész, The local time of iterated Brownian motion, J. Theoret. Probab. 9 (1996), 717–743. R. D. DeBlassie, Iterated Brownian motion in an open set, Ann. Appl. Prob. 14 (2004), no. 3, 1529-1558. Y. Hu, Hausdorff and packing measures of the level sets of iterated Brownian motion, J. Theoret. Probab. 12 (1999), 313– 346. Y. Hu, D. Pierre-Loti-Viaud, and Z. Shi, Laws of iterated logarithm for iterated Wiener proceses, J. Theoret. Probab. 8 (1995), 303–319. 50 D. Khoshnevisan, and T.M. Lewis, The uniform modulus of continuity of iterated Brownian motion. J. Theoret. Probab. 9 (1996), no. 2, 317–333. D. Khoshnevisan and T. M. Lewis, Chung’s law of the iterated logarithm for iterated Brownian motion, Ann. Inst. H. Poincaré Probab. Statist. 32 (1996), 349–359. D. Khoshnevisan and T. M. Lewis, Stochastic calculus for Brownian motion in a Brownian fracture, Ann. Appl. Probab. 9 (1999), 629–667. M.M. Meerschaert and H.P. Scheffler (2001) Limit Distributions for Sums of Independent Random Vectors: Heavy Tails in Theory and Practice. Wiley Interscience, New York. 51 M. M. Meerschaert and H. P. Scheffler, Limit theorems for continuous time random walks with infinite mean waiting times, J. Appl. Probab. 41 (2004), 623–638. M.M. Meerschaert, E. Nane and P. Vellaisamy (2008), Fractional Cauchy problems on bounded domains. Submitted E. Nane, Iterated Brownian motion in parabola-shaped domains, Potential Anal. 24 (2006), 105–123. E. Nane, Iterated Brownian motion in bounded domains in Rn, Stochastic Process. Appl. 116 (2006), 905–916. E. Nane, Higher order PDEŠs and iterated processes, Trans. Amer. Math. Soc. 360 (2008), 2681-2692. E. Nane, Laws of the iterated logarithm for α-time Brownian motion, Electron. J. Probab. 11 (2006), 434–459 (electronic). 52 E. Nane, Isoperimetric-type inequalities for iterated Brownian motion in Rn, Statist. Probab. Lett. 78 (2008), 90–95. E. Nane, Lifetime asymptotics of iterated Brownian motion in Rn, Esaim:PS, 11 (2007), 147–160. R. R. Nigmatullin, The realization of the generalized transfer in a medium with fractal geometry. Phys. Status Solidi B, 133 (1986), 425-430. E. Orsingher and L. Beghin (2004), Time-fractional telegraph equations and telegraph processes with Brownian time. Prob. Theory Rel. Fields 128, 141–160. E. Orsingher and L. Beghin (2008), Fractional diffusion equations and processes with randomly-varying time. Annal. Probab. (to appear). 53 Z. Shi and M. Yor, Integrability and lower limits of the local time of iterated Brownian motion, Studia Sci. Math. Hungar. 33 (1997), 279–298. Y. Xiao, Local times and related properties of multi-dimensional iterated Brownian motion, J. Theoret. Probab. 11 (1998), 383– 408. G. Zaslavsky, Fractional kinetic equation for Hamiltonian chaos. Chaotic advection, tracer dynamics and turbulent dispersion. Phys. D 76 (1994), 110-122. 54