GAUSSIAN MEASURES ON 1.1 BOREL MEASURES ON HILBERT SPACES CHAPTER 1

Size: px
Start display at page:

Download "GAUSSIAN MEASURES ON 1.1 BOREL MEASURES ON HILBERT SPACES CHAPTER 1"

Transcription

1 CAPTE GAUSSIAN MEASUES ON ILBET SPACES The aim of this chapter is to show the Minlos-Sazanov theorem and deduce a characterization of Gaussian measures on separable ilbert spaces by its Fourier transform. By using the notion of the ellinger integral we prove the Kakutani theorem on infinite product measures. As a consequence we obtain the Cameron-Martin theorem. For Gaussian measures on Banach spaces and their relationship with parabolic equations with many infinitely variables we refer to [] and [] and the references therein.. BOEL MEASUES ON ILBET SPACES Let be a real separable ilbert space, B the Borel σ-algebra on. Then B is a separable σ-algebra. A measure on the measurable space, B is called a Borel measure on. ere we only investigate finite Borel measures. Definition.. Let µ be a finite Borel measure on. The Fourier transform of µ is defined by µx : e i<x,y> µdy, x. Clearly µ possesses the following properties. Proposition.. The Fourier transform of a finite Borel measure satisfies the following properties

2 Gaussian measures on ilbert spaces µ0 µ. µ is continuous on. 3 µ is positive definite in the sense that l, µx l x k α l α k 0.. for any n, x, x,, x n, and α, α,, α n C. Proof: We have only to prove the third assertion. For n, x, x,..., x n, and α, α,..., α n C we have µx l x k α l α k e i<xl,y> e i<xk,y> α l α k µdy l, l, l, l e i<x l,y> α l e i<x k,y> α k µdy e i<xl, > α l, e i<xk, > α k L,µ e i<xk,y> α k µdy 0. ere L, µ denotes the space of all measurable functions f : satisfying fx µdx <. A natural question arises. Is any positive definite continuous functional on the Fourier transform of some finite Borel measure? The answer is affirmative if dim <. This is exactly the classical Bochner theorem see Theorem A..3. But in the infinite dimensional case the answer is negative. Take, for example, φx : exp x, x. Then it is easy to see that φ is a positive definite functional on. But φ is not the Fourier transform of any finite Borel measure on as we will see later see Proposition... To this end let us prove some auxiliary results. Lemma..3 Let φ be a positive definite functional on. x, y, Then, for any

3 . Borel measures on ilbert spaces 3 φx φ0, φx φ x. φx φy φ0 φ0 φx y. 3 φ0 φx φ0φ0 φx. Proof: For x, y, set φ0 φx A : φ x φ0 B : φ0 φx φy φ x φ0 φy x φ y φx y φ0 Since φ is positive definite, one can see that both A and B are positive definite matrices. In particular A t A. ence, φx φ x for all x. From deta 0, it follows that φx φ0. On the other hand, we have detb φ0 3 φ0 φx y φx[φ0φx φx yφy] + φy[φxφx y φ0φy] φ0 3 φ0 φx y φ0 φx φy + [φyφxφx y φ0]. Using the inequality a 3 ab a a b for b < a, we find Therefore, φ0 3 φ0 φx y φ0 φ0 φx y. 0 detb 4φ0 φ0 φx y φ0 φx φy This proves. Finally 3 follows from φ0 φx φ0 φx φ0 φx φ0 φ0φx + φx φ0 φ0φx. The following lemma will be useful for the proof of the Minlos-Sazanov theorem. Lemma..4 Let µ be a finite Borel measure on. Then the following assertions are equivalent.

4 4 Gaussian measures on ilbert spaces i x µdx <. ii There exists a positive, symmetric, trace class operator Q such that for x, y Qx, y x, z y, z µdz.. If ii holds, then TrQ x µdx. Proof: Suppose that ii holds. Let e n n N be an orthonormal basis of. Then x µdx x, e n µdx Qe n, e n TrQ <..3 n n Conversely, assume that i is satisfied. Thus, x, z y, z µdz x y z µdz. By the iesz representation theorem there exists Q L such that. is satisfied. Obviously, Q is positive and symmetric. Furthermore, by.3, TrQ x µdx <. ence Q is of trace class. Let show now the Minlos-Sazanov theorem. Theorem..5 Let φ be a positive definite functional on a separable real ilbert space. Then the following assertions are equivalent. φ is the Fourier transform of a finite Borel measure on. For every ε > 0 there is a symmetric positive operator of trace class Q ε such that Q ε x, x < φ0 φx < ε. 3 There exists a positive symmetric operator of trace class Q on such that φ is continuous or, equivalently, continuous at x 0 with respect to the semi-norm Q, where x Q : Qx, x Q / x, x. Proof: : Let φ µ. By applying the inequality cos ϑ ϑ, ϑ,

5 . Borel measures on ilbert spaces 5 we obtain, for any γ > 0, φ0 φx cos x, z µdz x, z µdz + µ {z : z > γ}. { z γ} Set µ A : µa { z γ} for A B, and apply Lemma..4 to µ. Thus there is a positive symmetric operator of trace class B γ such that B γ z, z { z γ} z, z z, z µdz. On the other hand, for every ε > 0 there is γ > 0 such that µ{z : z > γ} ε 4. Put Q ε : ε B γ, then φ0 φx ε Q εx, x + ε. : Assume that holds. Then φx is continuous at x 0. So, by Lemma..3-, φ is continuous on. Now, take any orthonormal basis e n n N of and for n put f i,,i n ω,, ω n : φω e + + ω n e n, ω j, j n. Then f i,,i n is a positive definite function on n. By the classical Bochner theorem see Theorem A..3 there exists a finite Borel measure µ i,,i n on n such that f i,,i n µ i,,i n. The family {µ i,,i n } satisfies the consistency conditions of Kolmogorov s extension theorem for measures cf. [30], p. 44. ence there is a unique finite Borel measure γ on, B such that µ i,,i n γ X i,, X in, where γ X i,, X in is defined by γ X i,, X in A γx i,, X in A for A B, and X j ω ω j, ω ω,, ω n,, j N. Claim: X k < γ-a.e.. Let P n be the standard Gaussian measure on n. Then e iay+ +anyn P n dy exp n a j.

6 6 Gaussian measures on ilbert spaces By assumption, we know that for every ε > 0 there is a positive symmetric operator Q ε of trace class such that ence, by Lemma..3-, Q ε x, x < φ0 φx < ε. φ0 φx ε + φ0 Q ε x, x for x. By Fubini s theorem we obtain ence, φ0 exp X k+j γdω φ0 γdω exp i y j X k+j P n dy n φ0 P n dy exp i y j X k+j γdω n φ0 P n dyφ y j e k+j n [φ0 φ y j e k+j ]P n dy n ε + φ0 Q ε y j e k+j, y l e k+l P n dy n l ε + φ0 Q ε e k+j, e l+j y j y l P n dy l, n ε + φ0 Q ε e k+j, e k+j yj P n dy } n {{} ε + φ0 Q ε e k+j, e k+j. φ0 exp X k+j γdω ε + φ0 Q ε e j, e j. jk+

7 . Borel measures on ilbert spaces 7 Now, let k, and ε 0, so we get lim exp X j γdω φ0 γ 0. k + jk+ This means that the function exp jk+ X j converges in L, γ to the constant function. Thus there is a subsequence of exp jk+ converging to γ a.e., which implies that and the claim is proved. Finally, let X j Xj < γ a.e., Xω : X j ωe j, ω. Then X is defined on γ-a.e., and X is an -valued measurable function. Put µ : γ X. Then µ is a finite Borel measure on and since µ i,,i n γ X i,, X in we obtain µ x, e j e j f,,n x, e,, x, e n φ x, e j e j. By letting n we obtain µ φ and the equivalence is proved. 3: Assume that holds. In take ε k for k N and λ k > 0 such that λ ktrq <. Set Q : k λ kq. It is obvious that Q is k a positive symmetric operator of trace class on. Moreover Q satisfies Qx, x < λ k Q x, x < k φ0 φx < k.

8 8 Gaussian measures on ilbert spaces So, by Lemma..3, φ is continuous on with respect to Q and hence 3 is proved. 3 : Conversely, suppose 3 is satisfied. So for every ε > 0 there is δ > 0 such that x Q < δ φ0 φx < ε. Set Q ε : δ Q. Then, Q ε x, x < φ0 φx < ε and Q ε satisfies.. GAUSSIAN MEASUES ON ILBET SPACES We will study a special class of Borel probability measures on. We first introduce the notions of mean vectors and covariance operators for general Borel probability measures. Definition.. Let µ be a Borel probability measure on. If for any x the function z x, z is integrable with respect to µ, and there exists an element m such that m, x x, z µdz, x, then m is called the mean vector of µ. If furthermore there is a positive symmetric linear operator B on such that Bx, y z m, x z m, y µdz, x, y, then B is called the covariance operator of µ. Mean vectors and covariance operators do not necessarily exist in general. But if x µdx <, then by iesz representation theorem, the mean vector m exists, and m x µdx. If furthermore, x µdx <, then by Lemma..4, there is a positive symmetric trace class operator Q such that Qx, y x, z y, z µdz x, y. Set Bx Qx m, x m, x. Then it is easy to verify that B is the covariance operator of µ. Note that B is also a positive symmetric trace class operator. We introduce now Gaussian measures.

9 . Gaussian measures on ilbert spaces 9 Definition.. Let µ be a Borel probability measure on. If for any x the random variable x, has a Gaussian distribution, then µ is called a Gaussian measure. emark..3 The scalar function x, has a Gaussian distribution means that there exists a real number m x and a positive number σ x such that µx e i x,z µdz exp im x σ x, x. In the sequel we will characterize Gaussian measures by means of Fourier transform. Lemma..4 Let α j j N such that α j. Then there exists a sequence of real numbers β j such that α j β j 0 for all j, βj < and α j β j. Proof: Set n 0 0 and define n k inductively as follows Then, n k. Put β j : α j k + n k : inf{l : n k+ jn k + α j l jn k + Then, for all j, α j β j 0, and βj n k+ α j }, k., n k + j n k+, k 0,,.... βj k0 jn k + k0 k + <, α j β j n k+ k0 jn k + k0 k0 k + α j β j k+ jn k + k +. α j The following result gives a characterization of Gaussian measures on separable ilbert spaces.

10 0 Gaussian measures on ilbert spaces Theorem..5 A Borel probability measure µ on is a Gaussian measure if and only if its Fourier transform is given by µx exp i < m, x > < Bx, x >, where m, B is a positive symmetric trace class operator on. In this case, m and B are the mean vector and covariance operator of µ respectively. Moreover, x µdx TrB + m. Proof: Let µ be a Gaussian measure on. Claim: x µdx <. By assumption, for any x, x, has a Gaussian distribution. Thus there are m x, and σ x > 0 such that µx e i<x,z> µdz exp im x σ x..4 Let e j be an orthonormal basis of. Since ξ m N m, σ dξ σ and ξn m, σ dξ m, we have x µdx x, e j µdx x jµdx j σe j + m e j. Let β j such that β j m ej 0 and β j <. Set ξx : β j e j, x By Schwarz inequality, the above series converges absolutely and ξx βj x, x. Moreover, ξ is linear. So by iesz representation theorem there is z such that ξx z, x, x. By assumption ξ z, is a Gaussian variable with a finite mean, i.e., β jm ej <. Now, by Lemma..4,

11 . Gaussian measures on ilbert spaces m e j <. Thus, in order to prove x µdx <, it suffices to check σ j <. By Theorem..5, there is a positive, symmetric, trace class operator Q such that Qx, x < µx < 6. ence, exp σ x µx Qx, x +, x..5 6 Without loss of generality we may assume that the kernel of Q is {0}. For x \ {0}, set y : 3<Qx,x> x. Then σ y 3 Qx, x σ x, and Qy, y 3. eplacing x by y in.5, we obtain exp σ x 6 Qx, x This implies that Thus, σ x 6 log 6 Qx, x, x. σe j 6 log 6TrQ <. ence, x µdx < and the claim is proved. So by the remark following Definition.. the mean vector m and the covariance operator B of µ exist. The above notation gives m x x, z µdz m, x and σ x From.4 we obtain µx exp x, z µdz m x [ x, z m, x ]µdz x, z m µdz Bx, x. i m, x Bx, x, x.

12 Gaussian measures on ilbert spaces Moreover, x µdx σe j + m e j TrB + m which proves the first implication. Conversely, let m and B be a positive, symmetric, trace class operator, and consider the positive definite functional φx exp i m, x Bx, x, x. Set Qx : Bx + m, x m, x. Then Q is a positive, symmetric, trace class operator on. Define Q on as follows x Q Q / x Qx, x / Bx, x + m, x /. Then φx is continuous at x 0 with respect to Q. So by Theorem..5, φ is the Fourier transform of some Borel probability measure µ on. Clearly for any x, x, is a Gaussian random variable with mean m, x and covariance Bx, x under µ. Thus, µ is a Gaussian measure. A Gaussian measure with mean vector m and covariance operator B will be denoted by N m, B. We propose now to compute some Gaussian integrals. Proposition..6 Let N 0, B be a Gaussian measure on. Then there is an orthonormal basis e n of such that Be n λ n e n, λ n 0, n N. Moreover, for any α < α 0 : inf n λ n, we have e α x N 0, Bdx αλ k deti αb. Proof: The first assertion follows from the fact that B is symmetric and positive. Since TrB λ k <, it follows that 0 αλ k < for α < α0. Furthermore, e α <x,e> N 0, Bdx e α ξ N 0, λ dξ πλ αλ. e α ξ e ξ λ dξ

13 . Gaussian measures on ilbert spaces 3 In similar way we have e α P n n <x,e k> N 0, Bdx αλ k and the result follows from the monotone convergence theorem. Before proving a more general result we propose first to study the transformation of a Gaussian measure by an affine mapping. Lemma..7 Let and be two separable ilbert spaces. Consider the affine transformation F : defined by F x Qx + z, where Q L, and z. If we set µ F : N m, B F, the measure defined by µ F A N m, BF A, A B, then µ F N Qm + z, QBQ. Proof: obtain Let compute the Fourier transform of µ F. From Theorem..5 we µ F x e i x,ey µ F dỹ e e i x,qy+z µdy e i x,z e i Q x,y µdy e i x,qm+z e QBQ x,x N Qm + z, QBQ x for x. So the lemma follows from Theorem..5. From the above lemma follows the following result. Proposition..8 Let α 0 : inf k λ k. Then, for any α < α 0, e α x N m, Bdx deti αb α exp I αb m, m.

14 4 Gaussian measures on ilbert spaces Proof: From Lemma..7 we have e α x N m, Bdx e α x+m N 0, Bdx e α m e α m πλk πλk e α m e πλk e α λ k α m k e πλk e α ξ +αm k ξ e ξ λ k dξ h i e αλk λ ξ αm k ξ k dξ λ k α m k αλ k m k αλ k αλ k αm k e αλ k deti αb e α I αb m,m. e αλ k λ ξ λ k αm k k αλ k dξ λk e ξ dξ αλ k Example..9 Let compute the integrals a x m N 0, Bdx, m N, b My N 0, Bdy, where M L. a For the integral in a we consider the function fα : e α x N 0, Bdx deti αb for α < α 0. Now, it is easy to see that, α 0 α deti αb is C and d dα deti αb TrBI αb deti αb, α < α 0. Furthermore we can differentiate under the integral sign. ence, x m N 0, Bdx m dm dα m deti αb α0.

15 . Gaussian measures on ilbert spaces 5 This implies that and b It follows from Lemma..7 that My N 0, Bdy x N 0, Bdx TrB x 4 N 0, Bdx TrB + TrB. y N 0, MBM dy. So by Theorem..5 we have My N 0, Bdy TrMBM TrM MB..6 By a same computation as above one has Proposition..0 For any α, m, we have e α,x N m, Bdx e α,m e Bα,α. We end this section by proving that the positive definite functional on defined by ϕx e x, x, is not the Fourier transform of any Borel measures provided that dim. Proposition.. Let Q be a positive, symmetric operator on a separable ilbert space. Then the functional φx exp < Qx, x >, x, is the Fourier transform of a probability measure on if and only if TrQ <. Proof: Suppose that TrQ <. Then φ0 and φ is Q -continuous positive functional on. So by Theorem..5 there exists a probability measure µ such that µx φx, x. To show the converse, assume that there is a probability measure µ such that e i<x,y> µdy exp < Qx, x >. Then by Theorem..5, for any ε 0, 3, there exists a positive, symmetric operator Q ε of trace class such that < Q ε x, x > < φ0 eφx < ε < Qx, x > < 3ε.

16 6 Gaussian measures on ilbert spaces Let now y 0 and < Q ε y 0, y 0 >: c, with c > 0. Let d > c arbitrary. Then y < Q 0 ε d, y0 c d > d <. ence, < Q y0 d, y0 d > < ε, i.e. < Qy 0, y 0 > < εd. Letting d c, we have < Qy 0, y 0 > ε < Q ε y 0, y 0 >. Since y 0 is arbitrary, we obtain < Qy, y > ε < Q ε y, y > for all y. In particular, for an orthonormal basis e n n N of, we obtain TrQ n < Qe n, e n > ε n < Q ε e n, e n > εtrq ε <. As an immediate consequence we obtain that the functional φx exp x, x, is not the Fourier transform of any probability measure on if dim..3 TE ELLINGE INTEGAL AND TE CAMEON-MATIN TEOEM The Cameron-Martin formula permits us to differentiate under the integral sign with respect to Gaussian measures in infinite dimensional ilbert spaces. This allows us to obtain some regularity results for parabolic equations with many infinitely variables. First we need some preparations. We denote by L + the space of all positive, symmetric operators of trace class on a separable ilbert space. Let B L + and consider an orthonormal basis e n n N of and a sequence λ n n N + such that Be n λ n e n, n N. Suppose also that ker B {0}. If we denote by x n :< x, e n >, then We set also Bx λ n x n e n and B x λ xn e n, x. n B n x : n λ k x k e k and B n x : Let consider, for a and n N, the function g a,n x : a, B n x λ k x ka k. λ k x ke k.

17 .3 The ellinger integral and the Cameron-Martin theorem 7 If a B then one can define the function g a x : λ k x ka k, x. The following proposition shows that it is always possible to define g a as an L, µ-function even if a B. Proposition.3. Let B L + with ker B {0} and µ : N 0, B its corresponding Gaussian measure on. Then the limit exists in L, µ. Moreover, for a given a. lim n + g a,n : g a g a x µdx a Proof: We have g a,n+p x g a,n x µdx n+p n+p kn+ h,kn+ n+p kn+ n+p kn+ λ k x ka k µdx λ h λ k ah a k λ k a k a k. x kµdx x h x k µdx ence g a,n n N is a Cauchy sequence in L, µ. Moreover, g a,n µdx a k λ k x kµdx a k and the theorem is proved by letting n. emark.3. Suppose that ker B {0} and take x such that B a, x 0 for all a. ence, B x 0 and so Bx 0, which implies that x 0. This proves that B is dense in. For the converse, let x with Bx 0. Thus, B x 0 and hence, B x, y x, B y 0 for

18 8 Gaussian measures on ilbert spaces all y. Since B, it follows that x 0. By the same arguments as in the proof of Proposition.3. one can show that g a is well defined as an L, µ function and g a L,µ a for a B. In the sequel we will use the notation g a x : a, B x, x. Proposition.3.3 Let B L + with ker B {0} and µ : N 0, B its corresponding Gaussian measure on. Then the limit lim n ega,n : e ga exists in L, µ for a given a. Moreover, for any a, e a,b x N 0, Bdx e a. Proof: By applying Proposition..0 we obtain e ga,n e ga,m µdx e B n a,x e B n a,x + B m a,x + e B m a,x µdx e P n a k + e P m a k e B n +B m a,x µdx e P n a k + e P m a k e P n a k + P m kn+ a k e P n a k + e P m kn+ a k e P m kn+ a k 0 n, m. This proves that e ga,n is a Cauchy sequence in L, µ and one can see that e a,b x N 0, Bdx e a is satisfied for every a. We propose now to recall the definition of the ellinger integral. Let µ, ν be two probability measures on a measurable space Ω, Σ. We say that µ and ν are singular notation: µ ν if there is a set B Σ such that µb 0 and νω \ B 0. It is easy to see that two probability measures µ and ν are singular if and only if for any ε > 0 there is B Σ such that µb < ε and νω \ B < ε. Further, µ is called ν-absolutely continuous notation: µ ν if νb 0 implies µb 0 for any B Σ. So by the theorem of adon-nikodym we know that if µ is ν-absolutely continuous,

19 .3 The ellinger integral and the Cameron-Martin theorem 9 then there is a non-negative measurable function ϕ defined on Ω, called the density function of µ, such that µb ϕωνdω for any B Σ. The density ϕ is denoted by B ϕω : dµ ω, ω Ω. dν If µ ν and ν µ are satisfied then µ and ν are called equivalent notation: µ ν. If µ ν, then the two density functions ϕ dµ dν dν and ψ dµ satisfy ϕωψω, a.e. ω Ω. ence, ϕω > 0 µ-a.e. ω Ω. Let now µ and ν two arbitrary probability measures on Ω, Σ. Let γ be a probability measure on Ω, Σ such that µ γ and ν γ. Such a measure exists, we have to take for example γ µ + ν. Thus, the following integral is well-defined µ, ν : Ω dµ dν ω ω γdω. dγ dγ This integral will be called the ellinger integral. Let now consider the measurable space, B, where B is the Borel field of subsets B of. On, B we consider two sequences of measures µ n and ν n with Then one has µ n, ν n Let us consider two infinite product measures µ n ν n, n N..7 n dν n dµ n x n µ n dx n. µ : µ n and ν : defined on, B. The following result is du to S. Kakutani [] and gives a condition under which these two measures µ and ν are equivalent. Theorem.3.4 Assume that.7 is satisfied. Then the following assertions hold. i If n µ n, ν n > 0 then µ ν and n ν n dν dµ x dν k x, a.e. x. dµ k

20 0 Gaussian measures on ilbert spaces ii If n µ n, ν n 0 then µ ν. Moreover, µ, ν µ n, ν n..8 n Proof: If we set ψ n x : n dν k dµ k x k for x and n N, then ψ n L,µ n dν k dµ k x k µdx n ν k dx k and ψ n ψ m L,µ n n dν k dµ k x k dν k dµ k x k m kn+ m kn+ dν k µ k, ν k m dν k x k µdx dµ k m dν k x k µdx dµ k kn+ x k µ k dx k dµ k.9 for any positive integers n and m with n < m. i If n µ n, ν n > 0 then lim m n,m kn+ µ k, ν k. ence, by.9, ψ n is a Cauchy sequence in L, µ and so there is ψ L, µ such that lim n ψ n ψ L,µ 0. Let prove now that ν µ and dν dµ x ψx, x, i.e. νb ψx µdx B

21 .3 The ellinger integral and the Cameron-Martin theorem for any B B. To this purpose it follows from ölder s inequality and.9 that ψ m x ψ n x µdx ψ m x + ψ n x µdx ψ m x ψ n x µdx 4 ψ m x ψ n x µdx m 8 µ k, ν k for n < m. Thus, kn+ lim n ψ n ψ L,µ 0. Finally let B B and set χ n x : χ B P n x, x, where χ B denotes the characteristic function of the measurable set B and P n x : x,..., x n, 0,.... So we have χ n x νdx χ B x,..., x n, 0,... ν dx... ν n dx n n n χ n x n dν k dµ k x k χ n xψ n x µdx. n µ k dx k Since ψn ψ in L, µ and by letting n we obtain νb ψx µdx. In a similar way one can see that µ ν. So we obtain µ ν. Finally, since µ ν, we have µ, ν ψx µdx lim ψ n x µdx n n dν k lim x k µ k dx k n dµ k lim n n µ k, ν k.

22 Gaussian measures on ilbert spaces So we obtain.8. ii If µ k, ν k 0 then for any ε > 0 there is n N such that n µ k, ν k < ε. Let B n B n with B n : {x,..., x n n : ψ n x,..., x n, 0,... Then, n µ k B n < B n n µ k dx n dν k dµ k x k > }. n ψ n x,..., x n, 0,... µ k dx B n n n dν k x k µ k dx B n dµ k n µ k, ν k < ε. By the same computation we obtain n ν k n \ B n n µ k, ν k < ε. Therefore, if we set B : B n kn+, then µb < ε and ν \ B < ε. This proves that µ ν. Suppose now that µ ν. Then there exists B B such that µb 0 and ν \ B 0. So by ölder s inequality, it follows that µ, ν B B dµ dγ dν x x γdx + dγ \B dµ x γdx dγ \B B dµ x γdx dγ dµ dγ dν x γdx dγ \B dν x x γdx dγ + dν x γdx dγ µb νb + µ \ B ν \ B 0. Therefore,.8 holds. This end the proof of the theorem. Let prove now the Cameron-Martin formula. We note here that the measure space, B can be identified with, B.

23 .3 The ellinger integral and the Cameron-Martin theorem 3 Corollary.3.5 Let B L + such that ker B {0} and µ : N 0, B and ν : N m, B be two Gaussian measures on, B. Then the following assertions hold. i The Gaussian measures µ and ν are equivalent if and only if m B. Moreover the adon-nikodym derivative is given by dν x exp dµ B m + < B x, B m >. ii The measures µ and ν are singular if and only if m B. Proof: We will apply Theorem.3.4 to the Gaussian measures µ and ν. To this purpose let compute the associated ellinger integral using.8. It follows from Proposition..0 that dν k µ k, ν k x k µ k dx k dµ k So by.8 we obtain This implies that k 4λ k e m e m k 8λ k. µ, ν µ, ν > 0 e m k x k λ k N 0, λ k dx k e m k 8λ k. m k λ k < m B. Moreover, in this case, it follows from Theorem.3.4 that dν dµ x dν k x dµ k exp e m k λ k e x k m k λ k B m + B x, B m, where x x ke k with x k : x, e k for an orthonormal basis e n of such that Be n λ n e n for n N. ere we used Proposition.3.3. Finally it is clear that the measures µ and ν are singular if and only if m B.

24 4 Gaussian measures on ilbert spaces Exercise.3.6 The Feldman-ajek theorem Let consider two linear operators B, B L + with ker B ker B {0} and an orthonormal basis e n of such that B e n λ n e n, n N, where λ n > 0 for all n N. On, B we consider the Gaussian measures µ : N 0, B and µ : N 0, B.. The commutative case: Suppose that B B B B. By using Theorem.3.4 show that a. if λ n α n n λ n+α n <, then µ µ. In this case dµ dµ x n b. if λ n α n n λ n+α n, then µ µ. exp λ n α n x, e n, λ n α n ere α n > 0, n N, are such that B e n α n e n, n N.. The General case: a Assume that there is S L + such that Show that µ µ. B B Id SB. b Assume that S L + and S <. Show that dµ x [deti S] exp dµ SI S B x, B x, x. ere L + is the set of positive ilbert-schmidt bounded linear operators on. That is, B L + if and only if B L, B positive and n Be n <.

5 Measure theory II. (or. lim. Prove the proposition. 5. For fixed F A and φ M define the restriction of φ on F by writing.

5 Measure theory II. (or. lim. Prove the proposition. 5. For fixed F A and φ M define the restriction of φ on F by writing. 5 Measure theory II 1. Charges (signed measures). Let (Ω, A) be a σ -algebra. A map φ: A R is called a charge, (or signed measure or σ -additive set function) if φ = φ(a j ) (5.1) A j for any disjoint

More information

HILBERT SPACES AND THE RADON-NIKODYM THEOREM. where the bar in the first equation denotes complex conjugation. In either case, for any x V define

HILBERT SPACES AND THE RADON-NIKODYM THEOREM. where the bar in the first equation denotes complex conjugation. In either case, for any x V define HILBERT SPACES AND THE RADON-NIKODYM THEOREM STEVEN P. LALLEY 1. DEFINITIONS Definition 1. A real inner product space is a real vector space V together with a symmetric, bilinear, positive-definite mapping,

More information

1 Inner Product Space

1 Inner Product Space Ch - Hilbert Space 1 4 Hilbert Space 1 Inner Product Space Let E be a complex vector space, a mapping (, ) : E E C is called an inner product on E if i) (x, x) 0 x E and (x, x) = 0 if and only if x = 0;

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

Notions such as convergent sequence and Cauchy sequence make sense for any metric space. Convergent Sequences are Cauchy

Notions such as convergent sequence and Cauchy sequence make sense for any metric space. Convergent Sequences are Cauchy Banach Spaces These notes provide an introduction to Banach spaces, which are complete normed vector spaces. For the purposes of these notes, all vector spaces are assumed to be over the real numbers.

More information

THEOREMS, ETC., FOR MATH 515

THEOREMS, ETC., FOR MATH 515 THEOREMS, ETC., FOR MATH 515 Proposition 1 (=comment on page 17). If A is an algebra, then any finite union or finite intersection of sets in A is also in A. Proposition 2 (=Proposition 1.1). For every

More information

Exercise Solutions to Functional Analysis

Exercise Solutions to Functional Analysis Exercise Solutions to Functional Analysis Note: References refer to M. Schechter, Principles of Functional Analysis Exersize that. Let φ,..., φ n be an orthonormal set in a Hilbert space H. Show n f n

More information

(U) =, if 0 U, 1 U, (U) = X, if 0 U, and 1 U. (U) = E, if 0 U, but 1 U. (U) = X \ E if 0 U, but 1 U. n=1 A n, then A M.

(U) =, if 0 U, 1 U, (U) = X, if 0 U, and 1 U. (U) = E, if 0 U, but 1 U. (U) = X \ E if 0 U, but 1 U. n=1 A n, then A M. 1. Abstract Integration The main reference for this section is Rudin s Real and Complex Analysis. The purpose of developing an abstract theory of integration is to emphasize the difference between the

More information

Tools from Lebesgue integration

Tools from Lebesgue integration Tools from Lebesgue integration E.P. van den Ban Fall 2005 Introduction In these notes we describe some of the basic tools from the theory of Lebesgue integration. Definitions and results will be given

More information

Lebesgue-Radon-Nikodym Theorem

Lebesgue-Radon-Nikodym Theorem Lebesgue-Radon-Nikodym Theorem Matt Rosenzweig 1 Lebesgue-Radon-Nikodym Theorem In what follows, (, A) will denote a measurable space. We begin with a review of signed measures. 1.1 Signed Measures Definition

More information

2 (Bonus). 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?

2 (Bonus). 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 9/5). Prove that every countable set A is measurable and µ(a) = 0. 2 (Bonus). Let A consist of points (x, y) such that either x or y is

More information

The Caratheodory Construction of Measures

The Caratheodory Construction of Measures Chapter 5 The Caratheodory Construction of Measures Recall how our construction of Lebesgue measure in Chapter 2 proceeded from an initial notion of the size of a very restricted class of subsets of R,

More information

Kernel Method: Data Analysis with Positive Definite Kernels

Kernel Method: Data Analysis with Positive Definite Kernels Kernel Method: Data Analysis with Positive Definite Kernels 2. Positive Definite Kernel and Reproducing Kernel Hilbert Space Kenji Fukumizu The Institute of Statistical Mathematics. Graduate University

More information

l(y j ) = 0 for all y j (1)

l(y j ) = 0 for all y j (1) Problem 1. The closed linear span of a subset {y j } of a normed vector space is defined as the intersection of all closed subspaces containing all y j and thus the smallest such subspace. 1 Show that

More information

Recall that if X is a compact metric space, C(X), the space of continuous (real-valued) functions on X, is a Banach space with the norm

Recall that if X is a compact metric space, C(X), the space of continuous (real-valued) functions on X, is a Banach space with the norm Chapter 13 Radon Measures Recall that if X is a compact metric space, C(X), the space of continuous (real-valued) functions on X, is a Banach space with the norm (13.1) f = sup x X f(x). We want to identify

More information

Integral Jensen inequality

Integral Jensen inequality Integral Jensen inequality Let us consider a convex set R d, and a convex function f : (, + ]. For any x,..., x n and λ,..., λ n with n λ i =, we have () f( n λ ix i ) n λ if(x i ). For a R d, let δ a

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

Elementary linear algebra

Elementary linear algebra Chapter 1 Elementary linear algebra 1.1 Vector spaces Vector spaces owe their importance to the fact that so many models arising in the solutions of specific problems turn out to be vector spaces. The

More information

Finite-dimensional spaces. C n is the space of n-tuples x = (x 1,..., x n ) of complex numbers. It is a Hilbert space with the inner product

Finite-dimensional spaces. C n is the space of n-tuples x = (x 1,..., x n ) of complex numbers. It is a Hilbert space with the inner product Chapter 4 Hilbert Spaces 4.1 Inner Product Spaces Inner Product Space. A complex vector space E is called an inner product space (or a pre-hilbert space, or a unitary space) if there is a mapping (, )

More information

Functional Analysis Winter 2018/2019

Functional Analysis Winter 2018/2019 Functional Analysis Winter 2018/2019 Peer Christian Kunstmann Karlsruher Institut für Technologie (KIT) Institut für Analysis Englerstr. 2, 76131 Karlsruhe e-mail: peer.kunstmann@kit.edu These lecture

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

Sobolev Spaces. Chapter 10

Sobolev Spaces. Chapter 10 Chapter 1 Sobolev Spaces We now define spaces H 1,p (R n ), known as Sobolev spaces. For u to belong to H 1,p (R n ), we require that u L p (R n ) and that u have weak derivatives of first order in L p

More information

08a. Operators on Hilbert spaces. 1. Boundedness, continuity, operator norms

08a. Operators on Hilbert spaces. 1. Boundedness, continuity, operator norms (February 24, 2017) 08a. Operators on Hilbert spaces Paul Garrett garrett@math.umn.edu http://www.math.umn.edu/ garrett/ [This document is http://www.math.umn.edu/ garrett/m/real/notes 2016-17/08a-ops

More information

1/12/05: sec 3.1 and my article: How good is the Lebesgue measure?, Math. Intelligencer 11(2) (1989),

1/12/05: sec 3.1 and my article: How good is the Lebesgue measure?, Math. Intelligencer 11(2) (1989), Real Analysis 2, Math 651, Spring 2005 April 26, 2005 1 Real Analysis 2, Math 651, Spring 2005 Krzysztof Chris Ciesielski 1/12/05: sec 3.1 and my article: How good is the Lebesgue measure?, Math. Intelligencer

More information

Analysis Comprehensive Exam Questions Fall 2008

Analysis Comprehensive Exam Questions Fall 2008 Analysis Comprehensive xam Questions Fall 28. (a) Let R be measurable with finite Lebesgue measure. Suppose that {f n } n N is a bounded sequence in L 2 () and there exists a function f such that f n (x)

More information

Fourier Transform & Sobolev Spaces

Fourier Transform & Sobolev Spaces Fourier Transform & Sobolev Spaces Michael Reiter, Arthur Schuster Summer Term 2008 Abstract We introduce the concept of weak derivative that allows us to define new interesting Hilbert spaces the Sobolev

More information

2 Lebesgue integration

2 Lebesgue integration 2 Lebesgue integration 1. Let (, A, µ) be a measure space. We will always assume that µ is complete, otherwise we first take its completion. The example to have in mind is the Lebesgue measure on R n,

More information

Real Analysis Problems

Real Analysis Problems Real Analysis Problems Cristian E. Gutiérrez September 14, 29 1 1 CONTINUITY 1 Continuity Problem 1.1 Let r n be the sequence of rational numbers and Prove that f(x) = 1. f is continuous on the irrationals.

More information

Measurable functions are approximately nice, even if look terrible.

Measurable functions are approximately nice, even if look terrible. Tel Aviv University, 2015 Functions of real variables 74 7 Approximation 7a A terrible integrable function........... 74 7b Approximation of sets................ 76 7c Approximation of functions............

More information

1 Fourier Integrals of finite measures.

1 Fourier Integrals of finite measures. 18.103 Fall 2013 1 Fourier Integrals of finite measures. Denote the space of finite, positive, measures on by M + () = {µ : µ is a positive measure on ; µ() < } Proposition 1 For µ M + (), we define the

More information

REAL AND COMPLEX ANALYSIS

REAL AND COMPLEX ANALYSIS REAL AND COMPLE ANALYSIS Third Edition Walter Rudin Professor of Mathematics University of Wisconsin, Madison Version 1.1 No rights reserved. Any part of this work can be reproduced or transmitted in any

More information

Signed Measures and Complex Measures

Signed Measures and Complex Measures Chapter 8 Signed Measures Complex Measures As opposed to the measures we have considered so far, a signed measure is allowed to take on both positive negative values. To be precise, if M is a σ-algebra

More information

9 Radon-Nikodym theorem and conditioning

9 Radon-Nikodym theorem and conditioning Tel Aviv University, 2015 Functions of real variables 93 9 Radon-Nikodym theorem and conditioning 9a Borel-Kolmogorov paradox............. 93 9b Radon-Nikodym theorem.............. 94 9c Conditioning.....................

More information

Wiener Measure and Brownian Motion

Wiener Measure and Brownian Motion Chapter 16 Wiener Measure and Brownian Motion Diffusion of particles is a product of their apparently random motion. The density u(t, x) of diffusing particles satisfies the diffusion equation (16.1) u

More information

4 Hilbert spaces. The proof of the Hilbert basis theorem is not mathematics, it is theology. Camille Jordan

4 Hilbert spaces. The proof of the Hilbert basis theorem is not mathematics, it is theology. Camille Jordan The proof of the Hilbert basis theorem is not mathematics, it is theology. Camille Jordan Wir müssen wissen, wir werden wissen. David Hilbert We now continue to study a special class of Banach spaces,

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

L p Functions. Given a measure space (X, µ) and a real number p [1, ), recall that the L p -norm of a measurable function f : X R is defined by

L p Functions. Given a measure space (X, µ) and a real number p [1, ), recall that the L p -norm of a measurable function f : X R is defined by L p Functions Given a measure space (, µ) and a real number p [, ), recall that the L p -norm of a measurable function f : R is defined by f p = ( ) /p f p dµ Note that the L p -norm of a function f may

More information

David Hilbert was old and partly deaf in the nineteen thirties. Yet being a diligent

David Hilbert was old and partly deaf in the nineteen thirties. Yet being a diligent Chapter 5 ddddd dddddd dddddddd ddddddd dddddddd ddddddd Hilbert Space The Euclidean norm is special among all norms defined in R n for being induced by the Euclidean inner product (the dot product). A

More information

Integration on Measure Spaces

Integration on Measure Spaces Chapter 3 Integration on Measure Spaces In this chapter we introduce the general notion of a measure on a space X, define the class of measurable functions, and define the integral, first on a class of

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

REAL ANALYSIS II HOMEWORK 3. Conway, Page 49

REAL ANALYSIS II HOMEWORK 3. Conway, Page 49 REAL ANALYSIS II HOMEWORK 3 CİHAN BAHRAN Conway, Page 49 3. Let K and k be as in Proposition 4.7 and suppose that k(x, y) k(y, x). Show that K is self-adjoint and if {µ n } are the eigenvalues of K, each

More information

Chapter 4. The dominated convergence theorem and applications

Chapter 4. The dominated convergence theorem and applications Chapter 4. The dominated convergence theorem and applications The Monotone Covergence theorem is one of a number of key theorems alllowing one to exchange limits and [Lebesgue] integrals (or derivatives

More information

Lectures 22-23: Conditional Expectations

Lectures 22-23: Conditional Expectations Lectures 22-23: Conditional Expectations 1.) Definitions Let X be an integrable random variable defined on a probability space (Ω, F 0, P ) and let F be a sub-σ-algebra of F 0. Then the conditional expectation

More information

Banach Spaces II: Elementary Banach Space Theory

Banach Spaces II: Elementary Banach Space Theory BS II c Gabriel Nagy Banach Spaces II: Elementary Banach Space Theory Notes from the Functional Analysis Course (Fall 07 - Spring 08) In this section we introduce Banach spaces and examine some of their

More information

MA359 Measure Theory

MA359 Measure Theory A359 easure Theory Thomas Reddington Usman Qureshi April 8, 204 Contents Real Line 3. Cantor set.................................................. 5 2 General easures 2 2. Product spaces...............................................

More information

Topological vectorspaces

Topological vectorspaces (July 25, 2011) Topological vectorspaces Paul Garrett garrett@math.umn.edu http://www.math.umn.edu/ garrett/ Natural non-fréchet spaces Topological vector spaces Quotients and linear maps More topological

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

It follows from the above inequalities that for c C 1

It follows from the above inequalities that for c C 1 3 Spaces L p 1. Appearance of normed spaces. In this part we fix a measure space (, A, µ) (we may assume that µ is complete), and consider the A - measurable functions on it. 2. For f L 1 (, µ) set f 1

More information

REAL ANALYSIS I Spring 2016 Product Measures

REAL ANALYSIS I Spring 2016 Product Measures REAL ANALSIS I Spring 216 Product Measures We assume that (, M, µ), (, N, ν) are σ- finite measure spaces. We want to provide the Cartesian product with a measure space structure in which all sets of the

More information

Measure Theory on Topological Spaces. Course: Prof. Tony Dorlas 2010 Typset: Cathal Ormond

Measure Theory on Topological Spaces. Course: Prof. Tony Dorlas 2010 Typset: Cathal Ormond Measure Theory on Topological Spaces Course: Prof. Tony Dorlas 2010 Typset: Cathal Ormond May 22, 2011 Contents 1 Introduction 2 1.1 The Riemann Integral........................................ 2 1.2 Measurable..............................................

More information

Chapter 1. Measure Spaces. 1.1 Algebras and σ algebras of sets Notation and preliminaries

Chapter 1. Measure Spaces. 1.1 Algebras and σ algebras of sets Notation and preliminaries Chapter 1 Measure Spaces 1.1 Algebras and σ algebras of sets 1.1.1 Notation and preliminaries We shall denote by X a nonempty set, by P(X) the set of all parts (i.e., subsets) of X, and by the empty set.

More information

. Then V l on K l, and so. e e 1.

. Then V l on K l, and so. e e 1. Sanov s Theorem Let E be a Polish space, and define L n : E n M E to be the empirical measure given by L n x = n n m= δ x m for x = x,..., x n E n. Given a µ M E, denote by µ n the distribution of L n

More information

Ergodic Theorems. Samy Tindel. Purdue University. Probability Theory 2 - MA 539. Taken from Probability: Theory and examples by R.

Ergodic Theorems. Samy Tindel. Purdue University. Probability Theory 2 - MA 539. Taken from Probability: Theory and examples by R. Ergodic Theorems Samy Tindel Purdue University Probability Theory 2 - MA 539 Taken from Probability: Theory and examples by R. Durrett Samy T. Ergodic theorems Probability Theory 1 / 92 Outline 1 Definitions

More information

2 n k In particular, using Stirling formula, we can calculate the asymptotic of obtaining heads exactly half of the time:

2 n k In particular, using Stirling formula, we can calculate the asymptotic of obtaining heads exactly half of the time: Chapter 1 Random Variables 1.1 Elementary Examples We will start with elementary and intuitive examples of probability. The most well-known example is that of a fair coin: if flipped, the probability of

More information

Weak Topologies, Reflexivity, Adjoint operators

Weak Topologies, Reflexivity, Adjoint operators Chapter 2 Weak Topologies, Reflexivity, Adjoint operators 2.1 Topological vector spaces and locally convex spaces Definition 2.1.1. [Topological Vector Spaces and Locally convex Spaces] Let E be a vector

More information

9. Banach algebras and C -algebras

9. Banach algebras and C -algebras matkt@imf.au.dk Institut for Matematiske Fag Det Naturvidenskabelige Fakultet Aarhus Universitet September 2005 We read in W. Rudin: Functional Analysis Based on parts of Chapter 10 and parts of Chapter

More information

Product measures, Tonelli s and Fubini s theorems For use in MAT4410, autumn 2017 Nadia S. Larsen. 17 November 2017.

Product measures, Tonelli s and Fubini s theorems For use in MAT4410, autumn 2017 Nadia S. Larsen. 17 November 2017. Product measures, Tonelli s and Fubini s theorems For use in MAT4410, autumn 017 Nadia S. Larsen 17 November 017. 1. Construction of the product measure The purpose of these notes is to prove the main

More information

Copyright 2010 Pearson Education, Inc. Publishing as Prentice Hall.

Copyright 2010 Pearson Education, Inc. Publishing as Prentice Hall. .1 Limits of Sequences. CHAPTER.1.0. a) True. If converges, then there is an M > 0 such that M. Choose by Archimedes an N N such that N > M/ε. Then n N implies /n M/n M/N < ε. b) False. = n does not converge,

More information

Riesz Representation Theorems

Riesz Representation Theorems Chapter 6 Riesz Representation Theorems 6.1 Dual Spaces Definition 6.1.1. Let V and W be vector spaces over R. We let L(V, W ) = {T : V W T is linear}. The space L(V, R) is denoted by V and elements of

More information

Homework I, Solutions

Homework I, Solutions Homework I, Solutions I: (15 points) Exercise on lower semi-continuity: Let X be a normed space and f : X R be a function. We say that f is lower semi - continuous at x 0 if for every ε > 0 there exists

More information

Signed Measures. Chapter Basic Properties of Signed Measures. 4.2 Jordan and Hahn Decompositions

Signed Measures. Chapter Basic Properties of Signed Measures. 4.2 Jordan and Hahn Decompositions Chapter 4 Signed Measures Up until now our measures have always assumed values that were greater than or equal to 0. In this chapter we will extend our definition to allow for both positive negative values.

More information

Functional Analysis. Martin Brokate. 1 Normed Spaces 2. 2 Hilbert Spaces The Principle of Uniform Boundedness 32

Functional Analysis. Martin Brokate. 1 Normed Spaces 2. 2 Hilbert Spaces The Principle of Uniform Boundedness 32 Functional Analysis Martin Brokate Contents 1 Normed Spaces 2 2 Hilbert Spaces 2 3 The Principle of Uniform Boundedness 32 4 Extension, Reflexivity, Separation 37 5 Compact subsets of C and L p 46 6 Weak

More information

Your first day at work MATH 806 (Fall 2015)

Your first day at work MATH 806 (Fall 2015) Your first day at work MATH 806 (Fall 2015) 1. Let X be a set (with no particular algebraic structure). A function d : X X R is called a metric on X (and then X is called a metric space) when d satisfies

More information

Chapter 8. General Countably Additive Set Functions. 8.1 Hahn Decomposition Theorem

Chapter 8. General Countably Additive Set Functions. 8.1 Hahn Decomposition Theorem Chapter 8 General Countably dditive Set Functions In Theorem 5.2.2 the reader saw that if f : X R is integrable on the measure space (X,, µ) then we can define a countably additive set function ν on by

More information

Mathematical foundations - linear algebra

Mathematical foundations - linear algebra Mathematical foundations - linear algebra Andrea Passerini passerini@disi.unitn.it Machine Learning Vector space Definition (over reals) A set X is called a vector space over IR if addition and scalar

More information

Sequences and Series of Functions

Sequences and Series of Functions Chapter 13 Sequences and Series of Functions These notes are based on the notes A Teacher s Guide to Calculus by Dr. Louis Talman. The treatment of power series that we find in most of today s elementary

More information

Real Analysis, 2nd Edition, G.B.Folland Elements of Functional Analysis

Real Analysis, 2nd Edition, G.B.Folland Elements of Functional Analysis Real Analysis, 2nd Edition, G.B.Folland Chapter 5 Elements of Functional Analysis Yung-Hsiang Huang 5.1 Normed Vector Spaces 1. Note for any x, y X and a, b K, x+y x + y and by ax b y x + b a x. 2. It

More information

Math 321 Final Examination April 1995 Notation used in this exam: N. (1) S N (f,x) = f(t)e int dt e inx.

Math 321 Final Examination April 1995 Notation used in this exam: N. (1) S N (f,x) = f(t)e int dt e inx. Math 321 Final Examination April 1995 Notation used in this exam: N 1 π (1) S N (f,x) = f(t)e int dt e inx. 2π n= N π (2) C(X, R) is the space of bounded real-valued functions on the metric space X, equipped

More information

If Y and Y 0 satisfy (1-2), then Y = Y 0 a.s.

If Y and Y 0 satisfy (1-2), then Y = Y 0 a.s. 20 6. CONDITIONAL EXPECTATION Having discussed at length the limit theory for sums of independent random variables we will now move on to deal with dependent random variables. An important tool in this

More information

1 Probability space and random variables

1 Probability space and random variables 1 Probability space and random variables As graduate level, we inevitably need to study probability based on measure theory. It obscures some intuitions in probability, but it also supplements our intuition,

More information

Chapter 4. Measure Theory. 1. Measure Spaces

Chapter 4. Measure Theory. 1. Measure Spaces Chapter 4. Measure Theory 1. Measure Spaces Let X be a nonempty set. A collection S of subsets of X is said to be an algebra on X if S has the following properties: 1. X S; 2. if A S, then A c S; 3. if

More information

Folland: Real Analysis, Chapter 7 Sébastien Picard

Folland: Real Analysis, Chapter 7 Sébastien Picard Folland: Real Analysis, Chapter 7 Sébastien Picard Problem 7.2 Let µ be a Radon measure on X. a. Let N be the union of all open U X such that µ(u) =. Then N is open and µ(n) =. The complement of N is called

More information

Real Variables # 10 : Hilbert Spaces II

Real Variables # 10 : Hilbert Spaces II randon ehring Real Variables # 0 : Hilbert Spaces II Exercise 20 For any sequence {f n } in H with f n = for all n, there exists f H and a subsequence {f nk } such that for all g H, one has lim (f n k,

More information

Lecture 4 Lebesgue spaces and inequalities

Lecture 4 Lebesgue spaces and inequalities Lecture 4: Lebesgue spaces and inequalities 1 of 10 Course: Theory of Probability I Term: Fall 2013 Instructor: Gordan Zitkovic Lecture 4 Lebesgue spaces and inequalities Lebesgue spaces We have seen how

More information

1 Math 241A-B Homework Problem List for F2015 and W2016

1 Math 241A-B Homework Problem List for F2015 and W2016 1 Math 241A-B Homework Problem List for F2015 W2016 1.1 Homework 1. Due Wednesday, October 7, 2015 Notation 1.1 Let U be any set, g be a positive function on U, Y be a normed space. For any f : U Y let

More information

INTRODUCTION TO MEASURE THEORY AND LEBESGUE INTEGRATION

INTRODUCTION TO MEASURE THEORY AND LEBESGUE INTEGRATION 1 INTRODUCTION TO MEASURE THEORY AND LEBESGUE INTEGRATION Eduard EMELYANOV Ankara TURKEY 2007 2 FOREWORD This book grew out of a one-semester course for graduate students that the author have taught at

More information

Analysis II Lecture notes

Analysis II Lecture notes Analysis II Lecture notes Christoph Thiele (lectures 11,12 by Roland Donninger lecture 22 by Diogo Oliveira e Silva) Summer term 2015 Universität Bonn July 5, 2016 Contents 1 Analysis in several variables

More information

TD 1: Hilbert Spaces and Applications

TD 1: Hilbert Spaces and Applications Université Paris-Dauphine Functional Analysis and PDEs Master MMD-MA 2017/2018 Generalities TD 1: Hilbert Spaces and Applications Exercise 1 (Generalized Parallelogram law). Let (H,, ) be a Hilbert space.

More information

3 Orthogonality and Fourier series

3 Orthogonality and Fourier series 3 Orthogonality and Fourier series We now turn to the concept of orthogonality which is a key concept in inner product spaces and Hilbert spaces. We start with some basic definitions. Definition 3.1. Let

More information

CHAPTER I THE RIESZ REPRESENTATION THEOREM

CHAPTER I THE RIESZ REPRESENTATION THEOREM CHAPTER I THE RIESZ REPRESENTATION THEOREM We begin our study by identifying certain special kinds of linear functionals on certain special vector spaces of functions. We describe these linear functionals

More information

MATH 650. THE RADON-NIKODYM THEOREM

MATH 650. THE RADON-NIKODYM THEOREM MATH 650. THE RADON-NIKODYM THEOREM This note presents two important theorems in Measure Theory, the Lebesgue Decomposition and Radon-Nikodym Theorem. They are not treated in the textbook. 1. Closed subspaces

More information

Elements of Positive Definite Kernel and Reproducing Kernel Hilbert Space

Elements of Positive Definite Kernel and Reproducing Kernel Hilbert Space Elements of Positive Definite Kernel and Reproducing Kernel Hilbert Space Statistical Inference with Reproducing Kernel Hilbert Space Kenji Fukumizu Institute of Statistical Mathematics, ROIS Department

More information

NORMS ON SPACE OF MATRICES

NORMS ON SPACE OF MATRICES NORMS ON SPACE OF MATRICES. Operator Norms on Space of linear maps Let A be an n n real matrix and x 0 be a vector in R n. We would like to use the Picard iteration method to solve for the following system

More information

THEOREMS, ETC., FOR MATH 516

THEOREMS, ETC., FOR MATH 516 THEOREMS, ETC., FOR MATH 516 Results labeled Theorem Ea.b.c (or Proposition Ea.b.c, etc.) refer to Theorem c from section a.b of Evans book (Partial Differential Equations). Proposition 1 (=Proposition

More information

Chapter 6. Integration. 1. Integrals of Nonnegative Functions. a j µ(e j ) (ca j )µ(e j ) = c X. and ψ =

Chapter 6. Integration. 1. Integrals of Nonnegative Functions. a j µ(e j ) (ca j )µ(e j ) = c X. and ψ = Chapter 6. Integration 1. Integrals of Nonnegative Functions Let (, S, µ) be a measure space. We denote by L + the set of all measurable functions from to [0, ]. Let φ be a simple function in L +. Suppose

More information

Functional Analysis Exercise Class

Functional Analysis Exercise Class Functional Analysis Exercise Class Week 9 November 13 November Deadline to hand in the homeworks: your exercise class on week 16 November 20 November Exercises (1) Show that if T B(X, Y ) and S B(Y, Z)

More information

Gaussian Random Fields

Gaussian Random Fields Gaussian Random Fields Mini-Course by Prof. Voijkan Jaksic Vincent Larochelle, Alexandre Tomberg May 9, 009 Review Defnition.. Let, F, P ) be a probability space. Random variables {X,..., X n } are called

More information

It follows from the above inequalities that for c C 1

It follows from the above inequalities that for c C 1 3 Spaces L p 1. In this part we fix a measure space (, A, µ) (we may assume that µ is complete), and consider the A -measurable functions on it. 2. For f L 1 (, µ) set f 1 = f L 1 = f L 1 (,µ) = f dµ.

More information

Real Analysis, 2nd Edition, G.B.Folland Signed Measures and Differentiation

Real Analysis, 2nd Edition, G.B.Folland Signed Measures and Differentiation Real Analysis, 2nd dition, G.B.Folland Chapter 3 Signed Measures and Differentiation Yung-Hsiang Huang 3. Signed Measures. Proof. The first part is proved by using addivitiy and consider F j = j j, 0 =.

More information

MATH 205C: STATIONARY PHASE LEMMA

MATH 205C: STATIONARY PHASE LEMMA MATH 205C: STATIONARY PHASE LEMMA For ω, consider an integral of the form I(ω) = e iωf(x) u(x) dx, where u Cc (R n ) complex valued, with support in a compact set K, and f C (R n ) real valued. Thus, I(ω)

More information

MATH5011 Real Analysis I. Exercise 1 Suggested Solution

MATH5011 Real Analysis I. Exercise 1 Suggested Solution MATH5011 Real Analysis I Exercise 1 Suggested Solution Notations in the notes are used. (1) Show that every open set in R can be written as a countable union of mutually disjoint open intervals. Hint:

More information

the convolution of f and g) given by

the convolution of f and g) given by 09:53 /5/2000 TOPIC Characteristic functions, cont d This lecture develops an inversion formula for recovering the density of a smooth random variable X from its characteristic function, and uses that

More information

Functional Analysis. Franck Sueur Metric spaces Definitions Completeness Compactness Separability...

Functional Analysis. Franck Sueur Metric spaces Definitions Completeness Compactness Separability... Functional Analysis Franck Sueur 2018-2019 Contents 1 Metric spaces 1 1.1 Definitions........................................ 1 1.2 Completeness...................................... 3 1.3 Compactness......................................

More information

A SHORT INTRODUCTION TO BANACH LATTICES AND

A SHORT INTRODUCTION TO BANACH LATTICES AND CHAPTER A SHORT INTRODUCTION TO BANACH LATTICES AND POSITIVE OPERATORS In tis capter we give a brief introduction to Banac lattices and positive operators. Most results of tis capter can be found, e.g.,

More information

Fonctions on bounded variations in Hilbert spaces

Fonctions on bounded variations in Hilbert spaces Fonctions on bounded variations in ilbert spaces Newton Institute, March 31, 2010 Introduction We recall that a function u : R n R is said to be of bounded variation (BV) if there exists an n-dimensional

More information

MAT 570 REAL ANALYSIS LECTURE NOTES. Contents. 1. Sets Functions Countability Axiom of choice Equivalence relations 9

MAT 570 REAL ANALYSIS LECTURE NOTES. Contents. 1. Sets Functions Countability Axiom of choice Equivalence relations 9 MAT 570 REAL ANALYSIS LECTURE NOTES PROFESSOR: JOHN QUIGG SEMESTER: FALL 204 Contents. Sets 2 2. Functions 5 3. Countability 7 4. Axiom of choice 8 5. Equivalence relations 9 6. Real numbers 9 7. Extended

More information

Operators with numerical range in a closed halfplane

Operators with numerical range in a closed halfplane Operators with numerical range in a closed halfplane Wai-Shun Cheung 1 Department of Mathematics, University of Hong Kong, Hong Kong, P. R. China. wshun@graduate.hku.hk Chi-Kwong Li 2 Department of Mathematics,

More information

MATH & MATH FUNCTIONS OF A REAL VARIABLE EXERCISES FALL 2015 & SPRING Scientia Imperii Decus et Tutamen 1

MATH & MATH FUNCTIONS OF A REAL VARIABLE EXERCISES FALL 2015 & SPRING Scientia Imperii Decus et Tutamen 1 MATH 5310.001 & MATH 5320.001 FUNCTIONS OF A REAL VARIABLE EXERCISES FALL 2015 & SPRING 2016 Scientia Imperii Decus et Tutamen 1 Robert R. Kallman University of North Texas Department of Mathematics 1155

More information

L p Spaces and Convexity

L p Spaces and Convexity L p Spaces and Convexity These notes largely follow the treatments in Royden, Real Analysis, and Rudin, Real & Complex Analysis. 1. Convex functions Let I R be an interval. For I open, we say a function

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