9 Measurable Functions and Their Properties

Total Page:16

File Type:pdf, Size:1020Kb

9 Measurable Functions and Their Properties 9 Measurable functions and their properties Throughout this section we assume that (X, ) is a measurable space. That is, we assumeX is a non-empty set and is aσ-algebraM inX. The sets in are called measurable subsets ofX. M M Notation (range, pre-image): SupposeY is a set andf:X Y is a function. The range off, denotedf(X), is the set of possible values taken by→ the functionf, i.e. f(X) := f(x):x X . { ∈ } GivenB Y , we define the pre-image (or inverse image) ofB underf as ⊂ 1 f − (B) := x X:f(x) B . { ∈ ∈ } Note that nothing is implied aboutf having an inverse function: in general, it will not 1 1 have one. For example,f − (B) could be empty, even ifB is not. The notationf − (B) is simply a shorthand for the set of points inX thatf maps into the setB. For example if 2 1 1 X=Y=R andf(x) =x thenf − ([1, 9]) = [1, 3] [ 3, 1] whilef − (( , 0)) =∅. ∪ − − −∞ Notation: Extended real line. We are mainly interested in functionsf fromX toR but sometimes it is useful to allowf to take values as well. We write R or ±∞ [ , ] to denote the setR ,+ , known as the extended real line. −∞ ∞ ∪{−∞ ∞} 1 Definition 9.1. We say that a functionf:X R is measurable, iff − ((α, ]) → ∞ ∈M for everyα R. ∈ If in factf(X) R, then we still use the same definition of measurability off; in this 1 ⊂ 1 case the setf − ((α, ]) will be the same asf − ((α, )). ∞ ∞ Remark. If we wish to clarify whichσ-algebra inX we mean (for example if more than oneσ-algebra is being considered inX, a situation common in probability theory), then we sayf is measurable with respect to (or for short, -measurable). M M Example 9.2(Indicator functions). LetE X. The function1 E :X R, defined by ⊂ → 1 ifx E; 1E(x) := ∈ 0 ifx E, � �∈ is called the indicator function or characteristic function ofE. The function1 E is a measurable function, if and only ifE (HW). ∈M Definition 9.3. SupposeW R is Borel (the setW could be all ofR), and let ⊂ := B W:B . BW { ⊂ ∈B} Then (W, W ) is a measurable space. A functiong:W R is called a Borel measur- able functionB , or just Borel function, if it is measurable→ with respect to . B W Lemma 9.4. (‘All continuous functions are Borel’). Suppose thatW R is open, ⊂ and thatf:W R is continuous. Thenf is Borel measurable on(W, W ). → B 28 Proof. Recall the continuity off is defined as follows: for allx W and allε> 0, there existsδ> 0 such that whenever∈ y x <δ then f(y) f(x) <ε. | − | | − | Letα R. Ifx W withf(x)>α, then we canfindδ> 0 such that for y x <δ we havey∈ W and∈f(y)>α. | − | ∈ Thereforef −((α, ]) =f −((α, )) W is open, and hence it is in W . Therefore,f is measurable on (W,∞ ). ∞ ⊂ B B W Lemma 9.5. SupposeY is a set andf:X Y is a function. Let := E Y: 1 → F { ⊂ f − (E) . Then is aσ-algebra inY. ∈M} F Proof. We leave this as HW. Theorem 9.6. Supposef:X R is a measurable function, andE is a Borel set inR. 1 → Thenf − (E) . ∈M 1 Proof. Set := E R:f − (E) . By Lemma 9.5, is aσ-algebra. Forα R F { ⊂ ∈M} F ∈ we have (α, ] by assumption, so that forα,β R withα<β we have that ∞ ∈F ∈ (α,β]=(α, ] (β, ] . Therefore (the bounded half-open intervals inR) so σ( )= ∞ by\ Theorem∞ ∈F 2.9. F⊃I F⊃ I B Theorem 9.7 (Composition of measurable functions). LetU R be an open set. If ⊂ f:X U is measurable, andg:U R is Borel (for example: if it is continuous), then → → h=g f, defined byh(x) =g(f(x)),h:X R, is measurable. ◦ → Proof. Letα R. We have: ∈ 1 h− ((α, ]) = x X:h(x)>α = x X:g(f(x))>α ∞ { ∈ } { ∈1 } = x X:f(x) g − ((α, ]) { ∈1 1 ∈ ∞ } =f − (g− ((α, ])). ∞ 1 Now,g − ((α, ]) sinceg is assumed Borel, and hence by Theorem 9.6, the set 1 1 ∞ ∈B f − (g− ((α, ])) is measurable (i.e., in ), as required. ∞ M Notation: Positive and negative parts. Fory R, we definey + := max(y, 0) ∈ (the positive part ofy) andy − := max( y, 0) (the negative part ofy). Note that + + − y=y y − and y =y +y −. Notation:− adding| | and multiplying functions. Suppose that we are given func- tionsf:X R,g:X R, and (forn N)f n :X R, and alsoa R. We → → + ∈ → ∈ define the functions f , f , f −, af, f+g andfg pointwise. That is, forx R we set | |+ + ∈ f (x) := f(x) , andf (x) := (f(x)) = max(f(x), 0), andf −(x) := max( f(x), 0), and (|af| )(x) :=| a(f|(x)), and (f+g)(x) :=f(x) +g(x) and (fg)(x) :=f(x)g(x). − Similarly we define the function supn fn pointwise, i.e. (supn fn)(x) = supn(fn(x)) for allx X. Likewise the functions lim sup n fn, infn fn, and lim infn fn and ∈ →∞ →∞ limn fn (if it exists) are defined pointwise, where we recall that for any sequence of →∞ numbers (an)n N we define ∈ lim sup an := lim (sup an, an+1, an+2,... ); n n { } →∞ →∞ lim inf an := lim (inf an, an+1, an+2,... ) n n →∞ →∞ { } 29 + + Remark 9.8. We note that f =f +f − andf=f f − (using pointwise addi- tion/subtraction both times).| | − + Corollary 9.9. Leta R. Iff:X R is measurable, so are af, f ,f andf −. ∈ → | | Proof. This follows from Theorem 9.7, takingg(y) = ay,g(y) = y ,g(y) =y + and | | g(y)=y − respectively. Theorem 9.10 (Sums and products of measurable functions). Iff:X R andg: → X R are measurable, then so aref+g andfg. → Proof. (ii) Supposef andg are measurable. Letα R. For any real u, v withu+v>α (so thatu>α v) there exists rationalq withu>q>∈ α v. Hence − − 1 (f+g) − ((α, ]) = x:f(x) +g(x)>α = q Q x:f(x)>q>α g(x) ∈ ∞ { } ∪ 1 { 1 − } = q Q[f − ((q, )) g − ((α q, ))] . ∪ ∈ ∞ ∩ − ∞ ∈M + Hencef+g is measurable. For the second part, note thatf=f f − so − + + + + + + fg=(f f −)(g g −) =f g +f −g− f g− g f −, − − − − so it suffices to provefg is measurable for the case withf 0 pointwise (i.e.f(X) [0, ) andg 0 pointwise. But in that case, forα> 0, iff(x)≥g(x)>α then there exists⊂ rational∞ q with≥ f(x)g(x)>q>α sof(x)>q>α/g(x), so x:f(x)g(x)>α = q Q (0, ) x:f(x)>q>α/g(x) , { } ∪ ∈ ∩ ∞ { }∈M and for the caseα = 0 we have x:f(x)g(x)>0 = n∞=1 x:f(x)g(x)>1/n , while forα< 0 we have x:f(x)g({x)>α =X }. ∪ { }∈M { } ∈M Remark 9.11. It can further be shown that iff:X [ , ] is measurable, then so + → −∞ ∞ aref := max f,0 andf − := max f,0 . Also, if f,g:X [0, ] are measurable, then so aref+g{ and}fg. (HW) {− } → ∞ Theorem 9.12 (Limits of measurable functions). Iff n :X R, are measurable func- → tions, defined forn N, then the functionsg:X R andh:X R defined by ∈ → → g := sup fn andh := lim sup fn n 1 n ≥ →∞ are also measurable. Similarly for infn 1 fn, and lim infn fn. ≥ →∞ Proof. For anyα R we have ∈ 1 g− ((α, ]) = x X : sup fn(x)>α ∞ { ∈ n 1 } ≥ = x X:f (x)>α for somen 1 { ∈ n ≥ } ∞ = x X:f (x)>α { ∈ n } n=1 � ∞ 1 = f − ((α, ]) . n ∞ ∈M n=1 , since � ∈M fn meas. � �� � 30 Thereforeg is measurable. Also g := inf n 1 fn = (sup n 1 fn) is also measurable. ≥ We can write − ≥ − h� = lim sup fk = inf sup fk, n k n n 1 k n →∞ ≥ ≥ ≥ and this is measurable by the previous paragraph. Similarly, lim infn fn = supn 1 infk n fk ≥ is also measurable. ≥ Corollary 9.13. Iff n :X R are measurable, andf(x) := lim n f(x) exists in R for →∞ eachx X, thenf is also measurable.→ ∈ For the proof of this, just note that iff(x) := lim n fn(x) exists, thenf = lim sup n fn. →∞ →∞ Definition 9.14. A functionf:X R is said to be simple, if (i) it is measurable and (ii) the rangef(X) is afinite set, i.e.→f takes onlyfinitely many values. Note 1: Here we exclude from the possible values. Note 2: It is convenient to±∞ include measurability in the definition of a simple function, though not all authors do so. Theorem 9.15 (Existence of Approximating Simple Functions). Letf:X [0, ] → ∞ be measurable. There exist nonnegative simple functionsf n,n N, such thatf n f pointwise, or in other words, such that for allx X: ∈ ↑ ∈ (a)0 f n(x) f n+1(x) for alln N; (b)f≤(x) f(x≤) asn . ∈ n → →∞ That is, every nonnegative measurable function can be expressed as an increasing limit of simple functions. Proof. For eachn 1, define the functionf n :X R by: ≥ → n n n n (k 1)10 − if (k 1)10 − f(x)<k10 − for some integer 1 k n10 ; fn(x) := − − ≤ ≤ ≤ n iff(x) n. � ≥ In other words,f n(x) is obtained by roundingf(x) down to thenth decimal place if f(x)<n, and settingf (x) =n iff(x) n.
Recommended publications
  • A Convenient Category for Higher-Order Probability Theory
    A Convenient Category for Higher-Order Probability Theory Chris Heunen Ohad Kammar Sam Staton Hongseok Yang University of Edinburgh, UK University of Oxford, UK University of Oxford, UK University of Oxford, UK Abstract—Higher-order probabilistic programming languages 1 (defquery Bayesian-linear-regression allow programmers to write sophisticated models in machine 2 let let sample normal learning and statistics in a succinct and structured way, but step ( [f ( [s ( ( 0.0 3.0)) 3 sample normal outside the standard measure-theoretic formalization of proba- b ( ( 0.0 3.0))] 4 fn + * bility theory. Programs may use both higher-order functions and ( [x] ( ( s x) b)))] continuous distributions, or even define a probability distribution 5 (observe (normal (f 1.0) 0.5) 2.5) on functions. But standard probability theory does not handle 6 (observe (normal (f 2.0) 0.5) 3.8) higher-order functions well: the category of measurable spaces 7 (observe (normal (f 3.0) 0.5) 4.5) is not cartesian closed. 8 (observe (normal (f 4.0) 0.5) 6.2) Here we introduce quasi-Borel spaces. We show that these 9 (observe (normal (f 5.0) 0.5) 8.0) spaces: form a new formalization of probability theory replacing 10 (predict :f f))) measurable spaces; form a cartesian closed category and so support higher-order functions; form a well-pointed category and so support good proof principles for equational reasoning; and support continuous probability distributions. We demonstrate the use of quasi-Borel spaces for higher-order functions and proba- bility by: showing that a well-known construction of probability theory involving random functions gains a cleaner expression; and generalizing de Finetti’s theorem, that is a crucial theorem in probability theory, to quasi-Borel spaces.
    [Show full text]
  • Jointly Measurable and Progressively Measurable Stochastic Processes
    Jointly measurable and progressively measurable stochastic processes Jordan Bell [email protected] Department of Mathematics, University of Toronto June 18, 2015 1 Jointly measurable stochastic processes d Let E = R with Borel E , let I = R≥0, which is a topological space with the subspace topology inherited from R, and let (Ω; F ;P ) be a probability space. For a stochastic process (Xt)t2I with state space E, we say that X is jointly measurable if the map (t; !) 7! Xt(!) is measurable BI ⊗ F ! E . For ! 2 Ω, the path t 7! Xt(!) is called left-continuous if for each t 2 I, Xs(!) ! Xt(!); s " t: We prove that if the paths of a stochastic process are left-continuous then the stochastic process is jointly measurable.1 Theorem 1. If X is a stochastic process with state space E and all the paths of X are left-continuous, then X is jointly measurable. Proof. For n ≥ 1, t 2 I, and ! 2 Ω, let 1 n X Xt (!) = 1[k2−n;(k+1)2−n)(t)Xk2−n (!): k=0 n Each X is measurable BI ⊗ F ! E : for B 2 E , 1 n [ −n −n f(t; !) 2 I × Ω: Xt (!) 2 Bg = [k2 ; (k + 1)2 ) × fXk2−n 2 Bg: k=0 −n −n Let t 2 I. For each n there is a unique kn for which t 2 [kn2 ; (kn +1)2 ), and n −n −n thus Xt (!) = Xkn2 (!). Furthermore, kn2 " t, and because s 7! Xs(!) is n −n left-continuous, Xkn2 (!) ! Xt(!).
    [Show full text]
  • Shape Analysis, Lebesgue Integration and Absolute Continuity Connections
    NISTIR 8217 Shape Analysis, Lebesgue Integration and Absolute Continuity Connections Javier Bernal This publication is available free of charge from: https://doi.org/10.6028/NIST.IR.8217 NISTIR 8217 Shape Analysis, Lebesgue Integration and Absolute Continuity Connections Javier Bernal Applied and Computational Mathematics Division Information Technology Laboratory This publication is available free of charge from: https://doi.org/10.6028/NIST.IR.8217 July 2018 INCLUDES UPDATES AS OF 07-18-2018; SEE APPENDIX U.S. Department of Commerce Wilbur L. Ross, Jr., Secretary National Institute of Standards and Technology Walter Copan, NIST Director and Undersecretary of Commerce for Standards and Technology ______________________________________________________________________________________________________ This Shape Analysis, Lebesgue Integration and publication Absolute Continuity Connections Javier Bernal is National Institute of Standards and Technology, available Gaithersburg, MD 20899, USA free of Abstract charge As shape analysis of the form presented in Srivastava and Klassen’s textbook “Functional and Shape Data Analysis” is intricately related to Lebesgue integration and absolute continuity, it is advantageous from: to have a good grasp of the latter two notions. Accordingly, in these notes we review basic concepts and results about Lebesgue integration https://doi.org/10.6028/NIST.IR.8217 and absolute continuity. In particular, we review fundamental results connecting them to each other and to the kind of shape analysis, or more generally, functional data analysis presented in the aforeme- tioned textbook, in the process shedding light on important aspects of all three notions. Many well-known results, especially most results about Lebesgue integration and some results about absolute conti- nuity, are presented without proofs.
    [Show full text]
  • LEBESGUE MEASURE and L2 SPACE. Contents 1. Measure Spaces 1 2. Lebesgue Integration 2 3. L2 Space 4 Acknowledgments 9 References
    LEBESGUE MEASURE AND L2 SPACE. ANNIE WANG Abstract. This paper begins with an introduction to measure spaces and the Lebesgue theory of measure and integration. Several important theorems regarding the Lebesgue integral are then developed. Finally, we prove the completeness of the L2(µ) space and show that it is a metric space, and a Hilbert space. Contents 1. Measure Spaces 1 2. Lebesgue Integration 2 3. L2 Space 4 Acknowledgments 9 References 9 1. Measure Spaces Definition 1.1. Suppose X is a set. Then X is said to be a measure space if there exists a σ-ring M (that is, M is a nonempty family of subsets of X closed under countable unions and under complements)of subsets of X and a non-negative countably additive set function µ (called a measure) defined on M . If X 2 M, then X is said to be a measurable space. For example, let X = Rp, M the collection of Lebesgue-measurable subsets of Rp, and µ the Lebesgue measure. Another measure space can be found by taking X to be the set of all positive integers, M the collection of all subsets of X, and µ(E) the number of elements of E. We will be interested only in a special case of the measure, the Lebesgue measure. The Lebesgue measure allows us to extend the notions of length and volume to more complicated sets. Definition 1.2. Let Rp be a p-dimensional Euclidean space . We denote an interval p of R by the set of points x = (x1; :::; xp) such that (1.3) ai ≤ xi ≤ bi (i = 1; : : : ; p) Definition 1.4.
    [Show full text]
  • 1 Measurable Functions
    36-752 Advanced Probability Overview Spring 2018 2. Measurable Functions, Random Variables, and Integration Instructor: Alessandro Rinaldo Associated reading: Sec 1.5 of Ash and Dol´eans-Dade; Sec 1.3 and 1.4 of Durrett. 1 Measurable Functions 1.1 Measurable functions Measurable functions are functions that we can integrate with respect to measures in much the same way that continuous functions can be integrated \dx". Recall that the Riemann integral of a continuous function f over a bounded interval is defined as a limit of sums of lengths of subintervals times values of f on the subintervals. The measure of a set generalizes the length while elements of the σ-field generalize the intervals. Recall that a real-valued function is continuous if and only if the inverse image of every open set is open. This generalizes to the inverse image of every measurable set being measurable. Definition 1 (Measurable Functions). Let (Ω; F) and (S; A) be measurable spaces. Let f :Ω ! S be a function that satisfies f −1(A) 2 F for each A 2 A. Then we say that f is F=A-measurable. If the σ-field’s are to be understood from context, we simply say that f is measurable. Example 2. Let F = 2Ω. Then every function from Ω to a set S is measurable no matter what A is. Example 3. Let A = f?;Sg. Then every function from a set Ω to S is measurable, no matter what F is. Proving that a function is measurable is facilitated by noticing that inverse image commutes with union, complement, and intersection.
    [Show full text]
  • Notes 2 : Measure-Theoretic Foundations II
    Notes 2 : Measure-theoretic foundations II Math 733-734: Theory of Probability Lecturer: Sebastien Roch References: [Wil91, Chapters 4-6, 8], [Dur10, Sections 1.4-1.7, 2.1]. 1 Independence 1.1 Definition of independence Let (Ω; F; P) be a probability space. DEF 2.1 (Independence) Sub-σ-algebras G1; G2;::: of F are independent for all Gi 2 Gi, i ≥ 1, and distinct i1; : : : ; in we have n Y P[Gi1 \···\ Gin ] = P[Gij ]: j=1 Specializing to events and random variables: DEF 2.2 (Independent RVs) RVs X1;X2;::: are independent if the σ-algebras σ(X1); σ(X2);::: are independent. DEF 2.3 (Independent Events) Events E1;E2;::: are independent if the σ-algebras c Ei = f;;Ei;Ei ; Ωg; i ≥ 1; are independent. The more familiar definitions are the following: THM 2.4 (Independent RVs: Familiar definition) RVs X, Y are independent if and only if for all x; y 2 R P[X ≤ x; Y ≤ y] = P[X ≤ x]P[Y ≤ y]: THM 2.5 (Independent events: Familiar definition) Events E1, E2 are indepen- dent if and only if P[E1 \ E2] = P[E1]P[E2]: 1 Lecture 2: Measure-theoretic foundations II 2 The proofs of these characterizations follows immediately from the following lemma. LEM 2.6 (Independence and π-systems) Suppose that G and H are sub-σ-algebras and that I and J are π-systems such that σ(I) = G; σ(J ) = H: Then G and H are independent if and only if I and J are, i.e., P[I \ J] = P[I]P[J]; 8I 2 I;J 2 J : Proof: Suppose I and J are independent.
    [Show full text]
  • (Measure Theory for Dummies) UWEE Technical Report Number UWEETR-2006-0008
    A Measure Theory Tutorial (Measure Theory for Dummies) Maya R. Gupta {gupta}@ee.washington.edu Dept of EE, University of Washington Seattle WA, 98195-2500 UWEE Technical Report Number UWEETR-2006-0008 May 2006 Department of Electrical Engineering University of Washington Box 352500 Seattle, Washington 98195-2500 PHN: (206) 543-2150 FAX: (206) 543-3842 URL: http://www.ee.washington.edu A Measure Theory Tutorial (Measure Theory for Dummies) Maya R. Gupta {gupta}@ee.washington.edu Dept of EE, University of Washington Seattle WA, 98195-2500 University of Washington, Dept. of EE, UWEETR-2006-0008 May 2006 Abstract This tutorial is an informal introduction to measure theory for people who are interested in reading papers that use measure theory. The tutorial assumes one has had at least a year of college-level calculus, some graduate level exposure to random processes, and familiarity with terms like “closed” and “open.” The focus is on the terms and ideas relevant to applied probability and information theory. There are no proofs and no exercises. Measure theory is a bit like grammar, many people communicate clearly without worrying about all the details, but the details do exist and for good reasons. There are a number of great texts that do measure theory justice. This is not one of them. Rather this is a hack way to get the basic ideas down so you can read through research papers and follow what’s going on. Hopefully, you’ll get curious and excited enough about the details to check out some of the references for a deeper understanding.
    [Show full text]
  • 1 Probability Measure and Random Variables
    1 Probability measure and random variables 1.1 Probability spaces and measures We will use the term experiment in a very general way to refer to some process that produces a random outcome. Definition 1. The set of possible outcomes is called the sample space. We will typically denote an individual outcome by ω and the sample space by Ω. Set notation: A B, A is a subset of B, means that every element of A is also in B. The union⊂ A B of A and B is the of all elements that are in A or B, including those that∪ are in both. The intersection A B of A and B is the set of all elements that are in both of A and B. ∩ n j=1Aj is the set of elements that are in at least one of the Aj. ∪n j=1Aj is the set of elements that are in all of the Aj. ∩∞ ∞ j=1Aj, j=1Aj are ... Two∩ sets A∪ and B are disjoint if A B = . denotes the empty set, the set with no elements. ∩ ∅ ∅ Complements: The complement of an event A, denoted Ac, is the set of outcomes (in Ω) which are not in A. Note that the book writes it as Ω A. De Morgan’s laws: \ (A B)c = Ac Bc ∪ ∩ (A B)c = Ac Bc ∩ ∪ c c ( Aj) = Aj j j [ \ c c ( Aj) = Aj j j \ [ (1) Definition 2. Let Ω be a sample space. A collection of subsets of Ω is a σ-field if F 1.
    [Show full text]
  • The Ito Integral
    Notes on the Itoˆ Calculus Steven P. Lalley May 15, 2012 1 Continuous-Time Processes: Progressive Measurability 1.1 Progressive Measurability Measurability issues are a bit like plumbing – a bit ugly, and most of the time you would prefer that it remains hidden from view, but sometimes it is unavoidable that you actually have to open it up and work on it. In the theory of continuous–time stochastic processes, measurability problems are usually more subtle than in discrete time, primarily because sets of measures 0 can add up to something significant when you put together uncountably many of them. In stochastic integration there is another issue: we must be sure that any stochastic processes Xt(!) that we try to integrate are jointly measurable in (t; !), so as to ensure that the integrals are themselves random variables (that is, measurable in !). Assume that (Ω; F;P ) is a fixed probability space, and that all random variables are F−measurable. A filtration is an increasing family F := fFtgt2J of sub-σ−algebras of F indexed by t 2 J, where J is an interval, usually J = [0; 1). Recall that a stochastic process fYtgt≥0 is said to be adapted to a filtration if for each t ≥ 0 the random variable Yt is Ft−measurable, and that a filtration is admissible for a Wiener process fWtgt≥0 if for each t > 0 the post-t increment process fWt+s − Wtgs≥0 is independent of Ft. Definition 1. A stochastic process fXtgt≥0 is said to be progressively measurable if for every T ≥ 0 it is, when viewed as a function X(t; !) on the product space ([0;T ])×Ω, measurable relative to the product σ−algebra B[0;T ] × FT .
    [Show full text]
  • Lebesgue Differentiation
    LEBESGUE DIFFERENTIATION ANUSH TSERUNYAN Throughout, we work in the Lebesgue measure space Rd;L(Rd);λ , where L(Rd) is the σ-algebra of Lebesgue-measurable sets and λ is the Lebesgue measure. 2 Rd For r > 0 and x , let Br(x) denote the open ball of radius r centered at x. The spaces 0 and 1 1. L Lloc The space L0 of measurable functions. By a measurable function we mean an L(Rd)- measurable function f : Rd ! R and we let L0(Rd) (or just L0) denote the vector space of all measurable functions modulo the usual equivalence relation =a:e: of a.e. equality. There are two natural notions of convergence (topologies) on L0: a.e. convergence and convergence in measure. As we know, these two notions are related, but neither implies the other. However, convergence in measure of a sequence implies a.e. convergence of a subsequence, which in many situations can be boosted to a.e. convergence of the entire sequence (recall Problem 3 of Midterm 2). Convergence in measure is captured by the following family of norm-like maps: for each α > 0 and f 2 L0, k k . · f α∗ .= α λ ∆α(f ) ; n o . 2 Rd j j k k where ∆α(f ) .= x : f (x) > α . Note that f α∗ can be infinite. 2 0 k k 6 k k Observation 1 (Chebyshev’s inequality). For any α > 0 and any f L , f α∗ f 1. 1 2 Rd The space Lloc of locally integrable functions. Differentiation at a point x is a local notion as it only depends on the values of the function in some open neighborhood of x, so, in this context, it is natural to work with a class of functions defined via a local property, as opposed to global (e.g.
    [Show full text]
  • HARMONIC ANALYSIS 1. Maximal Function for a Locally Integrable
    HARMONIC ANALYSIS PIOTR HAJLASZ 1. Maximal function 1 n For a locally integrable function f 2 Lloc(R ) the Hardy-Littlewood max- imal function is defined by Z n Mf(x) = sup jf(y)j dy; x 2 R : r>0 B(x;r) The operator M is not linear but it is subadditive. We say that an operator T from a space of measurable functions into a space of measurable functions is subadditive if jT (f1 + f2)(x)j ≤ jT f1(x)j + jT f2(x)j a.e. and jT (kf)(x)j = jkjjT f(x)j for k 2 C. The following integrability result, known also as the maximal theorem, plays a fundamental role in many areas of mathematical analysis. p n Theorem 1.1 (Hardy-Littlewood-Wiener). If f 2 L (R ), 1 ≤ p ≤ 1, then Mf < 1 a.e. Moreover 1 n (a) For f 2 L (R ) 5n Z (1.1) jfx : Mf(x) > tgj ≤ jfj for all t > 0. t Rn p n p n (b) If f 2 L (R ), 1 < p ≤ 1, then Mf 2 L (R ) and p 1=p kMfk ≤ 2 · 5n=p kfk for 1 < p < 1, p p − 1 p kMfk1 ≤ kfk1 : Date: March 28, 2012. 1 2 PIOTR HAJLASZ The estimate (1.1) is called weak type estimate. 1 n 1 n Note that if f 2 L (R ) is a nonzero function, then Mf 62 L (R ). Indeed, R if λ = B(0;R) jfj > 0, then for jxj > R we have Z λ Mf(x) ≥ jfj ≥ n ; B(x;R+jxj) !n(R + jxj) n and the function on the right hand side is not integrable on R .
    [Show full text]
  • 1 Random Vectors and Product Spaces
    36-752 Advanced Probability Overview Spring 2018 4. Product Spaces Instructor: Alessandro Rinaldo Associated reading: Sec 2.6 and 2.7 of Ash and Dol´eans-Dade;Sec 1.7 and A.3 of Durrett. 1 Random Vectors and Product Spaces We have already defined random variables and random quantities. A special case of the latter and generalization of the former is a random vector. Definition 1 (Random Vector). Let (Ω; F;P ) be a probability space. Let X :Ω ! IRk be a measurable function. Then X is called a random vector . There arises, in this definition, the question of what σ-field of subsets of IRk should be used. When left unstated, we always assume that the σ-field of subsets of a multidimensional real space is the Borel σ-field, namely the smallest σ-field containing the open sets. However, because IRk is also a product set of k sets, each of which already has a natural σ-field associated with it, we might try to use a σ-field that corresponds to that product in some way. 1.1 Product Spaces The set IRk has a topology in its own right, but it also happens to be a product set. Each of the factors in the product comes with its own σ-field. There is a way of constructing σ-field’s of subsets of product sets directly without appealing to any additional structure that they might have. Definition 2 (Product σ-Field). Let (Ω1; F1) and (Ω2; F2) be measurable spaces. Let F1 ⊗ F2 be the smallest σ-field of subsets of Ω1 × Ω2 containing all sets of the form A1 × A2 where Ai 2 Fi for i = 1; 2.
    [Show full text]