Scribd Logo
Upload

Welcome to Scribd!

  • Convergence Inlp

    Uploaded by

    Belsty Wale Kibret
    13 views5 pages

    Original Title

    convergence inLp

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    13 views5 pages

    Convergence Inlp

    Uploaded by

    Belsty Wale Kibret

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 5
    Consergence in ie 204 B Lebesgue Measure and Integral B.7 Convergence Theorems Review Tf (Fn}ea1 is @ sequence of measurable functions that converge pointwise almost everywhere to a function f, then it need not be true that littn soo fi fn will converge to fp f (see Exercise B.57 below). In this section we review several important theorems related to this sue. ‘The following result is also known as the Beppo Levi Theorem, We say that, a sequence of extended real-valued functions (Ja}nen i8 monotone increasing if file) < fal2) <--- for all x. We vite f/f to denote that (fa}aen is ‘monotone increasing and that fa(2) > f(2) pointwise ‘Theorem B.53 (Monotone Convergence Theorem). Let {fa}ncn be @ monotone increasing sequence of measurable, nonnegative functions on a measurable set E CRY, and define f(z) = littn soo fa(2), 80 fy 7 f. Then din fie for Proof. Note that R(fi,E) C R(fz,E) C +. and that R(f,E) = UR( fa, E). Exercise B.10(A) therefore implies that [R(Ju, B)] converges to |R(Y,E)). O Since changing the value of a function on a set of zero measure does not change the value of its integral, it suffices to assume that the hypotheses in the Monotone Convergence Theorem hold ae. instead of everywhere, and the same is true of the other theoreras in this section. As an application of the Monotone Convergence Theorem, we ean now complete the proof of Theorem B.35. The proof relies on the useful fact that wwe can always find simple functions that Increase polutwise to an arbitrary nonnegative measurable function. Exercise B.54. Suppose that B CR and J, g: D ~ [0,00] are measurable. (2) Show thet i a S$ F() < ge Fb, én(2) = ny iff@)2n is a simple function, én(z) / f(z) for each x, and if f is bounded then n converges uniformly to J. Apply the Monotone Convergence 'heorcm to conclude that fyén 7 fpf. (&) Prove Theorem B.35, ie., show that fe(f-+9) = Jp t+ fps. Sinco the partial sums of a series of nonnegative functions form a mono- tone increasing sequence, we have another application of the Monotone Con- vergence Theorem, this time to the problem of interchanging a sum and an integral, This is a spocial case of the abstract version of Tonelil’s Theorem, compare Theorems B.65 and D.70. jew B.7 Convergence Theorems 205 Corollary B.55. Let {fn}nen be a sequence of measuroble, nonnegative furtc- ‘Hons on a measurable set EC R#. Then LQ") - Zhe Ifthe fimetions f, aro nonnogative but are not monotone increasing, then we may not be able to interchange a limit and an integral. However, the following result states that if the functions f,, are all nonnegative, then we do at loost have @ particular inoquality. Exercise B.56 (Fatou’s Lemma). If {fe nen 18 a sequence of measurable, nonnegative functions on measurable set BC R¢, then [Cunt to) < image ff AC inti Exercise B.57. Give examples showing that sit inequality can hold in Fa to's Lemma [Now we come to the workhorse ofthe stable of convergence theorems. Exercise B.58 (Lebesgue Dominated Convergence Theorem). As- sume {fa}ncn is a sequence of measurable functions on measurable set EC R®, either extended real-valued or complex-valued, such that: (2) £(2) = limnn soo fal) exists for ac. 2 € E, and (b) there exists g € L3(B) such that | fu(2)| < o(2) ac. for every n. ‘Then fy, converges to f in L'norm, ic., Bim Sus = i, [f= fol and, consequently, Tet us make some general observations concerning convergence in L?(B). By definition, if p is finite then fn — f in L¥(E) means that lin, f te) ~ Ale ae Hence, there will surely be a need to use the convergence theorems presented above when dealing with convergence in L?, but most likely applied to |f— fal? rather than to Ja and f. In the & spaces, convergence in £ norm implies componentwise conver~ agence (cco Problem A.5). The situation in L?(R) is a litle different, For p= 00, if fy — f in L™(B) then it follows that fu(z) — f(z) pointwise ae. However, far p finite, an L?-convergent sequence need not converge pointwise. 206 © B Lebesgue Measure and Integral Exercise B.59. Let 0

    0, then there exists a bounded open set U 2 B such thet |U\B) < « ‘Also, by Problem B.4, there exists a compact set KC B such that |E\K] <« By Urysohn's Lemma (Theorem A.103), we can find a continuous function a: Rt — Ré such that 0 < 6 <1, @= 1 on K, and 9 = 0 on RU. Then 8 € C.(R4), and we have xe — 012 = [2-9 = fen” < Ki 0, then there exists a 5 > 0 such that if ECR is moasurable and |B <6, then fx |f| <¢. In particular, if |B} = 0, then B.18. Show that L7(R) is separable for 1 < p< 00, but L(R) is not sepa rable. Problem Hinks B.IT Hint: Define B, = (x € R¢: [f(2)| f in DRS). B.8 Hint: The set of all intervals (a,)) with rational endpoiats is countable. Show that 346 Solutions S= {Soaxwean nN Re (ra), Im(rx) €@, ax

    You might also like