Q
stringlengths
4
3.96k
A
stringlengths
1
3k
Result
stringclasses
4 values
Corollary 5.4.6. A subset \( E \) of \( \mathbb{Z} \) is a spectral set for \( M \) if and only if \( k\left( E\right) \) is contained in the closed linear span of all functions \( {e}_{n} \), where \( n \in \mathbb{Z} \smallsetminus E \) .
Proof. Because \( M \) is semisimple and regular, \( E \) is a spectral set if and only if \( k\left( E\right) = \overline{j\left( E\right) } \) . The statement now follows immediately from the preceding lemma.
No
Corollary 5.4.7. Let \( E \subseteq \mathbb{Z} \) and \( m \in \mathbb{Z} \) . Then \( E \) is a spectral set for \( M \) if and only if \( E + m = \{ n + m : n \in E\} \) is a spectral set for \( M \) .
Proof. Of course, it suffices to show that if \( E \) is a spectral set, then so is \( E + m \) . Let \( f \in k\left( {E + m}\right) \) and let \( g \in M \) be defined by \( g\left( t\right) = f\left( t\right) {e}^{-{imt}} \) . Then, for each \( n \in E \) ,\n\n\[{\varphi }_{n}\left( g\right) = \frac{1}{2\pi }{\int }...
Yes
Theorem 5.4.8. Let \( E \) and \( F \) be subsets of \( \mathbb{Z} \) such that \( E \subseteq F \) and \( F \smallsetminus E \) is finite. Then \( E \) is a set of synthesis for \( M \) if and only if \( F \) is a set of synthesis for \( M \) .
Proof. Suppose first that \( E \) is a spectral set. To show that \( F \) is a spectral set, proceeding inductively, it suffices to treat the case where \( F = E \cup \{ m\} \) for some \( m \in \mathbb{Z} \smallsetminus E \) . Let \( f \in k\left( F\right) \) and \( \epsilon > 0 \) be given. There exists \( g \in j\le...
Yes
Theorem 5.4.9. Let \( E \) and \( F \) be subsets of \( \mathbb{Z} \) such that \( F \subseteq E \). (i) If \( F \) is a Ditkin set, so is \( E \). (ii) If \( E \) is a Ditkin set and \( E \smallsetminus F \) is finite, then \( F \) is a Ditkin set.
Proof. (i) Because, by Theorem 5.2.2, a closed countable union of Ditkin sets is a Ditkin set, it suffices to show that if \( F \) is a Ditkin set and \( m \in \mathbb{Z} \smallsetminus F \), then \( E = F \cup \{ m\} \) is a Ditkin set.\n\nLet \( f \in k\left( E\right) \) and \( \epsilon > 0 \) be given. Since \( F \)...
Yes
Lemma 5.4.10. For a subset \( E \) of \( \mathbb{Z} \), the following conditions are equivalent.\n\n(i) \( E \) is a spectral set for \( M \) .\n\n(ii) \( \langle \mu, f\rangle = 0 \) for every \( f \in k\left( E\right) \) and every \( \mu \in {M}^{ * } \) with \( \sigma \left( \mu \right) \subseteq E \) .
Proof. (i) \( \Rightarrow \) (ii) Let \( f \in k\left( E\right) \) and \( \mu \in {M}^{ * } \) with \( \sigma \left( \mu \right) \subseteq E \) . Because \( E \) is a spectral set, by Lemma 5.4.5 \( k\left( E\right) \) is the closed linear span of functions \( {e}_{n}, n \in \mathbb{Z} \smallsetminus E \) . Thus there ...
Yes
Proposition 5.4.11. For a subset \( E \) of \( \mathbb{Z} \) the following conditions are equivalent.\n\n(i) \( E \) is a spectral for \( M \) .\n\n(ii) If \( \mu \in {M}^{ * } \) is such that \( \sigma \left( \mu \right) \subseteq E \), then \( \mu \) is the \( {w}^{ * } \) -limit of functionals in the linear span of ...
Proof. (i) \( \Rightarrow \) (ii) Let \( \mu \in {M}^{ * } \) such that \( \sigma \left( \mu \right) \subseteq E \) . Because \( E \) is a spectral set, Lemma 5.4.10 implies that \( \langle \mu, f\rangle = 0 \) for every \( f \in k\left( E\right) \) . Towards a contradiction, assume that \( \mu \) does not belong to th...
Yes
Theorem 5.4.12. Every subset of \( 4\mathbb{Z} \) is a spectral set for \( M \) .
Proof. Let \( E \) be a nonempty subset of \( 4\mathbb{Z} \) and \( \nu = {gdx} + \mu \in {M}^{ * } \) with \( \sigma \left( \nu \right) \subseteq E \) . Let \( p \) be any trigonometric polynomial and let \( q\left( t\right) = p\left( {t + \pi /2}\right) \) .\n\nThen\n\[ \widehat{q}\left( n\right) = {e}^{{in\pi }/2}{\...
Yes
Lemma 5.4.13. Let \( f \in M \) be defined by\n\n\[ f\left( t\right) = \left\{ \begin{array}{rr} 1 & \text{ if }\left| t\right| \leq \pi /2 \\ - 1 & \text{ if }\pi /2 < \left| t\right| \leq \pi \end{array}\right. \]\n\nThen \( \widehat{f}\left( 0\right) = 0 \) and, for \( n \in \mathbb{Z} \smallsetminus \{ 0\} \) ,\n\n...
Proof. For \( n \in \mathbb{Z}, n \neq 0 \), we have\n\n\[ \widehat{f}\left( n\right) = \frac{1}{2\pi }\left( {-{\int }_{-\pi }^{-\pi /2}{e}^{-{int}}{dt} + {\int }_{-\pi /2}^{\pi /2}{e}^{-{int}}{dt} - {\int }_{\pi /2}^{\pi }{e}^{-{int}}{dt}}\right) \]\n\n\[ = \frac{1}{2\pi in}\left( {{e}^{{in\pi }/2} - {e}^{in\pi } - {...
Yes
Lemma 5.4.14. Let \( \mu = {\delta }_{-\pi /2} + {\delta }_{\pi /2} \in {M}^{ * } \) . Then\n\n\[ \widehat{\mu }\left( n\right) = 2\cos \left( \frac{n\pi }{2}\right) \]\n\nfor every \( n \in \mathbb{Z} \) . In particular, \( \sigma \left( \mu \right) = 2\mathbb{Z} \) .
Proof. For each \( n \in \mathbb{Z} \) ,\n\n\[ \widehat{\mu }\left( n\right) = {\int }_{\mathbb{T}}{e}^{-{int}}{d\mu }\left( t\right) = {e}^{{in\pi }/2} + {e}^{-{in\pi }/2} = 2\cos \left( \frac{n\pi }{2}\right) . \]\n\nSince \( \cos \left( {{n\pi }/2}\right) \neq 0 \) if and only if \( n \in 2\mathbb{Z} \), we get that...
Yes
Corollary 5.4.15. Let \( f \) and \( \mu \) be as in Lemmas 5.4.13 and 5.4.14, respectively. Then \( \langle \mu, f - p * f\rangle = 2 \) for every trigonometric polynomial \( p \) .
Proof. Notice first that \( \langle \mu, f\rangle = f\left( {-\pi /2}\right) + f\left( {\pi /2}\right) = 2 \) . Then\n\n\[ \langle \mu, f - p * f\rangle = 2 - \langle \mu, p * f\rangle = 2 - \mathop{\sum }\limits_{{n \in \mathbb{Z}}}\widehat{p}\left( n\right) \widehat{f}\left( n\right) \widehat{\mu }\left( n\right) \]\...
Yes
Corollary 5.4.16. The union of two disjoint sets of synthesis for \( M \) need not be a set of synthesis. For example, \( 2\mathbb{Z} = 4\mathbb{Z} \cup \left( {4\mathbb{Z} + 2}\right) \) fails to be of synthesis even though both \( 4\mathbb{Z} \) and \( 4\mathbb{Z} + 2 \) are sets of synthesis.
Proof. By Theorem 5.4.12, \( 4\mathbb{Z} \) is a spectral set and hence so is \( 4\mathbb{Z} + 2 \) by Corollary 5.4.7. However, \( 2\mathbb{Z} \) is not of synthesis. Indeed, let \( f \) and \( \mu \) be as in Lemmas 5.4.13 and 5.4.14. Then \( f \in k\left( {2\mathbb{Z}}\right) \) and \( \sigma \left( \mu \right) = 2\...
Yes
Theorem 5.4.17. Let \( E \) be a finite subset of \( \mathbb{Z} = \Delta \left( M\right) \) . Then \( E \) is a set of synthesis, but \( E \) fails to be a Ditkin set.
Proof. We first show that \( E \) is a set of synthesis. Since the trigonometric polynomials \( p \) are dense in \( M \) (Lemma 5.4.2) and \( \widehat{p} \) has finite support for any such \( p,\varnothing \) is a spectral set. Thus let \( E \) be nonempty and let \( f \in k\left( E\right) \) and \( \mu \in {M}^{ * } ...
Yes
Theorem 5.5.1. Let \( G \) be a locally compact Abelian group.\n\n(i) \( {L}^{1}\left( G\right) \) satisfies Ditkin’s condition at infinity.\n\n(ii) Let \( K \) be a compact subset of \( \widehat{G} \) and \( \epsilon > 0 \) . Then there exists \( f \in {L}^{1}\left( G\right) \) such that \( \widehat{f} = 1 \) on \( K,...
Proof. (i) Because \( {L}^{1}\left( G\right) \) has an approximate identity, it suffices to show that \( {L}^{1}\left( G\right) \) is Tauberian. Let \( F \) denote the space of all functions \( f \in {L}^{2}\left( G\right) \) such that \( \widehat{f} \) equals almost everywhere a continuous function with compact suppor...
Yes
Corollary 5.5.2. Every proper closed ideal of \( {L}^{1}\left( G\right) \) is contained in some maximal modular ideal.
Because \( \Delta \left( {{L}^{1}\left( G\right) }\right) = \widehat{G} \) is discrete when \( G \) is compact, Theorem 5.5.1 and Corollary 5.2.15 imply the following
No
Corollary 5.5.4. Let \( f \in {L}^{1}\left( G\right) \) and let \( I \) denote the closed linear subspace of \( {L}^{1}\left( G\right) \) generated by all the translates \( {L}_{x}f, x \in G \) . Then the following conditions are equivalent.\n\n(i) \( I = {L}^{1}\left( G\right) \).\n\n(ii \( \widehat{f}\left( \alpha \r...
Proof. Suppose that \( I \neq {L}^{1}\left( G\right) \) . Then, by Proposition 1.4.7, \( I \) is a proper closed ideal of \( {L}^{1}\left( G\right) \) and hence, by Corollary 5.5.2, there exists a maximal modular ideal \( M \) of \( {L}^{1}\left( G\right) \) such that \( M \supseteq I \) . By Theorem 2.7.2, \( M \) is ...
Yes
Lemma 5.5.6. Given \( f \in {L}^{1}\left( G\right) \) and \( \epsilon > 0 \), there exists \( g \in {\mathcal{F}}_{C}\left( G\right) \) such that \[ {\begin{Vmatrix}f * g - \widehat{f}\left( {1}_{G}\right) g\end{Vmatrix}}_{1} < \epsilon \] and hence also \[ \parallel f * g{\parallel }_{1} < C\left| {\widehat{f}\left( {...
Proof. For any \( g \in {L}^{1}\left( G\right) \), we have \[ {\begin{Vmatrix}f * g - \widehat{f}\left( {1}_{G}\right) g\end{Vmatrix}}_{1} \leq {\int }_{G}\left| {f\left( y\right) }\right| \cdot {\begin{Vmatrix}{L}_{y}g - g\end{Vmatrix}}_{1}{dy}. \] Choose a compact subset \( K \) of \( G \) such that \[ {\int }_{G \sm...
Yes
Lemma 5.5.7. Let \( H \) be a closed subgroup of \( G \) and let \( f \in {L}^{1}\left( G\right) \) and \( \epsilon > 0 \) . Then there exists a measure \( \mu \in M\left( G\right) \) with the following properties.\n\n(i) \( \parallel \mu \parallel < 1 + \epsilon \) and \( \widehat{\mu } = 1 \) on a neighbourhood of \(...
Proof. We first establish the lemma for \( f \in {C}_{c}\left( G\right) \) . Choose \( C > 1 \) such that \( C < 1 + \epsilon \) and \( \left( {C - 1}\right) {\begin{Vmatrix}{T}_{H}f\end{Vmatrix}}_{1} < \epsilon /2 \) . By Lemma 5.5.6 there exists \( h \in {L}^{1}\left( H\right) \) such that \( \parallel h{\parallel }_...
Yes
Theorem 5.5.8. Let \( G \) be a locally compact Abelian group, \( \Gamma \) a closed subgroup of \( \widehat{G} \), and \( \alpha \in \widehat{G} \). For every \( \delta > 0, k\left( {\alpha \Gamma }\right) \) possesses an approximate identity of norm bound \( 2 + \delta \) which is contained in \( j\left( {\alpha \Gam...
Proof. By the remark preceding the theorem, we can assume that \( \alpha = {1}_{G} \). The closed subgroup \( \Gamma \) of \( \widehat{G} \) is of the form \( \Gamma = \widehat{G/H} \), where \( H \) is the annihilator of \( \Gamma \) in \( \widehat{G} \); that is, \( H = \{ x \in G : \gamma \left( x\right) = 1 \) for ...
Yes
Theorem 5.5.9. Let \( G \) be a locally compact Abelian group, \( H \) a closed subgroup of \( G \), and embed \( \widehat{G/H} \) into \( \widehat{G} \) . Let \( E \) be a closed subset of \( \widehat{G/H} \) . (i) \( E \) is a set of synthesis for \( {L}^{1}\left( {G/H}\right) \) if and only if \( E \) is a set of sy...
Proof. It follows from Theorem 5.2.6(i) that if \( E \) is a set of synthesis (respectively, Ditkin set) for \( {L}^{1}\left( G\right) \), then \( E \) is a set of synthesis (respectively, Ditkin set) for \( {L}^{1}\left( {G/H}\right) = {L}^{1}\left( G\right) /\ker {T}_{H} \) . Moreover, the converse conclusion of (i) ...
Yes
Corollary 5.5.10. Let \( {\Gamma }_{1},\ldots ,{\Gamma }_{m} \) be closed subgroups of \( \widehat{G} \) and let \( {\gamma }_{1},\ldots ,{\gamma }_{m} \) be characters of \( G \) . For each \( j = 1,\ldots, m \), let \( {m}_{j} \in {\mathbb{N}}_{0} \) and for \( 1 \leq k \leq {m}_{j} \) , let \( {\Delta }_{jk} \) be a...
Proof. Finite unions of Ditkin sets are Ditkin sets (Lemma 5.2.1). Therefore we can assume that \( m = 1 \) . Moreover, if \( F \) is a Ditkin set then so is \( {\gamma F} \) for every \( \gamma \in \widehat{G} \) . Thus we can further assume that \( E \) is of the form \( E = \) \( \Gamma \smallsetminus \mathop{\bigcu...
Yes
Corollary 5.6.2. Let \( G \) be a compact Abelian group and \( I \) a closed ideal of \( {L}^{1}\left( G\right) \) . Then \( I \) has a bounded approximate identity if and only if \( I = \mu * {L}^{1}\left( G\right) \) for some idempotent measure \( \mu \in M\left( G\right) \) .
Proof. Let \( I \) have a bounded approximate identity. Then, by Proposition 5.6.1, there exists an idempotent measure \( \mu \) on \( b\left( G\right) = G \) such that \( \widehat{\mu } \) equals the characteristic function of \( \widehat{G} \smallsetminus h\left( I\right) \) . Now,\n\n\[ h\left( {\mu * {L}^{1}\left( ...
Yes
Proposition 5.6.3. If \( \mu \in F\left( G\right) \), then the support group of \( \mu \), the smallest closed subgroup of \( G \) whose complement is a \( \mu \) -null set, is compact.
Proof. Assume that \( \mu \neq 0 \) and let \( H \) be the support group of \( \mu \) . Then \( \mu \) can be regarded as an element of \( F\left( H\right) \) . We show that \( \widehat{H} \) is discrete and hence \( H \) is compact.\n\nLet \( \gamma \) be any nontrivial character of \( H \) . We claim that \( \gamma \...
Yes
Proposition 5.6.8. Let \( G \) be a compact Abelian group and \( \mu \in F\left( G\right) \) . Then the set\n\n\[ S\left( \mu \right) = \{ \alpha \in \widehat{G} : \widehat{\mu }\left( \alpha \right) \neq 0\} \]\n\nbelongs to \( \mathcal{R}\left( \widehat{G}\right) \) .
Proof. For \( k \in {\mathbb{Z}}^{ * } = \mathbb{Z} \smallsetminus \{ 0\} \), let \( S{\left( \mu \right) }_{k} = \{ \alpha \in \widehat{G} : \widehat{\mu }\left( \alpha \right) = k\} \) . Since the range of \( \widehat{\mu } \) is finite, it suffices to show that \( S{\left( \mu \right) }_{k} \in \mathcal{R}\left( \wi...
Yes
Lemma 5.7.1. Let \( E \) and \( F \) be closed subsets of \( \Delta \left( A\right) \) and \( \Delta \left( B\right) \), respectively, and let \( \psi \in F \) . Then\n\n\[{\phi }_{\psi }\left( {j\left( {E \times F}\right) }\right) \subseteq j\left( E\right) .
Proof. Let \( x \in j\left( {E \times F}\right) \) . There exists a compact subset \( C \) of \( \Delta \left( {A\widehat{ \otimes }B}\right) = \) \( \Delta \left( A\right) \times \Delta \left( B\right) \) such that \( \widehat{x} = 0 \) outside of \( C \) and \( C \cap \left( {E \times F}\right) = \varnothing \) . Now...
Yes
Theorem 5.7.2. Let \( A \) and \( B \) be regular commutative Banach algebras and suppose that \( A{\widehat{ \otimes }}_{\pi }B \) is semisimple. Let \( E \) and \( F \) be closed subsets of \( \Delta \left( A\right) \) and \( \Delta \left( B\right) \), respectively. If \( E \times F \) is a set of synthesis (Ditkin s...
Proof. Let \( a \in A \) and \( \epsilon > 0 \) be given. Choose \( \psi \in F \) and \( b \in B \) such that \( \psi \left( b\right) = 1 \) . If \( E \times F \) is a set of synthesis, there exists \( x \in j\left( {E \times F}\right) \) such that \( \parallel x - a \otimes b\parallel \leq \epsilon \) . Then \( {\phi ...
Yes
Corollary 5.7.4. Let \( G \) be a compact Abelian group and \( A \) a semisimple and regular commutative Banach algebra. Suppose that \( A \) is Tauberian and has a bounded approximate identity. Then spectral synthesis holds for \( {L}^{1}\left( {G, A}\right) \) if and only if it holds for \( A \) .
Proof. Recall first that \( {L}^{1}\left( {G, A}\right) = {L}^{1}\left( G\right) {\widehat{ \otimes }}_{\pi }A \) is semisimple by Theorem 2.11.6 because \( {L}^{1} \) -spaces share the approximation property. Thus, if spectral synthesis holds for \( {L}^{1}\left( {G, A}\right) \) then it holds for \( A \) (Theorem 5.7...
Yes
Lemma 5.7.5. Suppose that \( A \) and \( B \) are Tauberian commutative Banach algebras, and let \( E \) and \( F \) be closed subsets of \( \Delta \left( A\right) \) and \( \Delta \left( B\right) \), respectively. Then\n\n\[ j\left( E\right) \otimes B + A \otimes j\left( F\right) \subseteq \overline{j\left( {E \times ...
Proof. It suffices to show that if\n\n\[ x = \mathop{\sum }\limits_{{k = 1}}^{n}{a}_{k} \otimes {b}_{k} + \mathop{\sum }\limits_{{l = 1}}^{m}{c}_{l} \otimes {d}_{l} \]\n\nwhere \( {a}_{k} \in j\left( E\right) ,{b}_{k} \in B,{c}_{l} \in A \), and \( {d}_{l} \in j\left( F\right) \), then \( x \in \overline{j\left( {E \ti...
Yes
Lemma 5.7.6. Let \( F \) be a closed subset of \( \Delta \left( A\right) \) and \( \varphi \in \Delta \left( A\right) \). (i) Every \( x \in k\left( {\{ \varphi \} \times F}\right) \) has a representation of the form \[ x = \mathop{\sum }\limits_{{j = 1}}^{\infty }\left( {{a}_{j} \otimes {b}_{j}}\right) + e \otimes b \...
Proof. (i) Let \( x \in k\left( {\{ \varphi \} \times F}\right) \). Then \( \mathop{\sum }\limits_{{j = 1}}^{\infty }{y}_{j} \otimes {b}_{j} \), where \( {y}_{j} \in A,{b}_{j} \in B \), and \( \mathop{\sum }\limits_{{j = 1}}^{\infty }\begin{Vmatrix}{y}_{j}\end{Vmatrix} \cdot \begin{Vmatrix}{b}_{j}\end{Vmatrix} < \infty...
Yes
Corollary 5.7.7. Let \( A \) and \( B \) be regular and Tauberian commutative Banach algebras and suppose that \( A{\widehat{ \otimes }}_{\pi }B \) is semisimple. Let \( \{ \varphi \} \) be a spectral set for \( A \) and \( F \subseteq \Delta \left( B\right) \) a spectral set for \( B \) . Then \( \{ \varphi \} \times ...
Proof. It follows from part (ii) of Lemma 5.7.6 and the hypotheses that\n\n\[ k\left( {\{ \varphi \} \times F}\right) = k\left( \varphi \right) {\widehat{ \otimes }}_{\pi }B + A \otimes k\left( F\right) \]\n\n\[ = \overline{j\left( \varphi \right) }{\widehat{ \otimes }}_{\pi }B + A \otimes \overline{j\left( F\right) } ...
Yes
Theorem 5.7.8. Let \( A \) and \( B \) be unital regular commutative Banach algebras. Suppose that \( A{\widehat{ \otimes }}_{\pi }B \) is semisimple, and let \( E \) be a closed subset of \( \Delta \left( {A{\widehat{ \otimes }}_{\pi }B}\right) \) such that the set\n\n\[ \n\{ \varphi \in \Delta \left( A\right) : \left...
Proof. We apply Theorem 5.2.13 to \( A{\widehat{ \otimes }}_{\pi }B, T = \Delta \left( A\right) \) and the projection\n\n\[ \n\phi : \Delta \left( {A{\widehat{ \otimes }}_{\pi }B}\right) = \Delta \left( A\right) \times \Delta \left( B\right) \rightarrow \Delta \left( A\right) ,\;\left( {\varphi ,\psi }\right) \rightarr...
Yes
Corollary 5.7.9. Let \( X \) be a compact Hausdorff space and \( A \) a regular and semisimple commutative Banach algebra with identity. Suppose that \( \Delta \left( A\right) \) is scattered and that every singleton \( \{ \varphi \} ,\varphi \in \Delta \left( A\right) \), is a spectral set and satisfies condition (D)....
Proof. We only have to note that \( A{\widehat{ \otimes }}_{\pi }C\left( X\right) \) is semisimple because \( A \) and \( C\left( X\right) \) are semisimple and \( C\left( X\right) \) has the approximation property.
Yes
(ii) If \( A = \mathop{\bigcup }\limits_{n}{A}_{n} \), then\n\n\[{m}^{ * }\left( A\right) \leq \mathop{\sum }\limits_{n}{m}^{ * }\left( {A}_{n}\right)\]
Proof. (ii) We can, and do, assume that \( \mathop{\sum }\limits_{n}{m}^{ * }\left( {A}_{n}\right) < \infty \) . With this in mind, let \( \epsilon > 0 \) be given, and choose for each \( n \) a sequence \( \left( {I}_{n, j}\right) \) of intervals that cover \( {A}_{n} \) and satisfy\n\n\[ \mathop{\sum }\limits_{j}\ope...
Yes
(i) If \( {F}_{1} \) and \( {F}_{2} \) are disjoint closed bounded sets, then\n\n\[ \n{m}^{ * }\left( {{F}_{1} \cup {F}_{2}}\right) = {m}^{ * }\left( {F}_{1}\right) + {m}^{ * }\left( {F}_{2}\right) \n\]
Proof. (i) Let \( \delta > 0 \) be chosen so that no interval of diameter less than \( \delta \) meets both \( {F}_{1} \) and \( {F}_{2} \) (e.g., \( \delta < \frac{1}{2}d\left( {{F}_{1},{F}_{2}}\right) \), where \( d\left( {{F}_{1},{F}_{2}}\right) \) is the distance between the disjoint closed sets \( {F}_{1} \) and \...
Yes
(i) If \( A \) and \( B \) are measurable, then so is \( A \cap B \) .
Proof. (i) Pick \( {F}_{A},{F}_{B} \) closed and \( {G}_{A},{G}_{B} \) open such that\n\n\[ {F}_{A} \subseteq A \subseteq {G}_{A}\text{ and }{m}^{ * }\left( {{G}_{A} \smallsetminus {F}_{A}}\right) \leq \frac{\epsilon }{2} \]\n\n\[ {F}_{B} \subseteq B \subseteq {G}_{B}\text{ and }{m}^{ * }\left( {{G}_{B} \smallsetminus ...
Yes
Theorem 1.7 (The Brunn-Minkowski Theorem). Let \( n \geq 1 \), and let \( {\lambda }_{n} \) denote Lebesgue measure on \( {\mathbb{R}}^{n} \). If \( A, B \), and \( A + B \) are measurable subsets of \( {\mathbb{R}}^{n} \), then\n\n\[{\left( {\lambda }_{n}\left( A + B\right) \right) }^{1/n} \geq {\left( {\lambda }_{n}\...
Proof. To prove (BM), suppose that each of \( A \) and \( B \) is the union of finitely many rectangles whose interiors are disjoint. We proceed by induction on the total number of rectangles in \( A \) and \( B \). It is important to realize that the inequality is unaffected if we translate \( A \) and \( B \) indepen...
Yes
Theorem 1.8. For any Borel set \( B \subseteq {\mathbb{R}}^{n} \) we have\n\n\[ \n{\lambda }_{n}\left( B\right) \leq {\left( \frac{\operatorname{diam}B}{2}\right) }^{n}{\lambda }_{n}\left( {B}_{{\ell }_{n}^{2}}\right) \n\]\n\nwhere as usual \( {B}_{{\ell }_{n}^{2}} \) denotes the closed unit ball of \( {\mathbb{R}}^{n}...
Proof. Without loss of sleep we can assume that\n\n\[ \nd \mathrel{\text{:=}} \operatorname{diam}B < \infty \text{.} \n\]\n\nThis in mind, realize that if \( x, y, \in B \), then \( \parallel x - y\parallel \leq d \) . Hence\n\n\[ \nB - B \subseteq d{B}_{{\ell }_{n}^{2}} \n\]\n\nand so\n\n\[ \n2{\lambda }_{n}{\left( B\...
Yes
Theorem 1.9 (Vitali). Let \( \mathcal{F} \) be a family of closed balls in \( {\mathbb{R}}^{n} \) that covers a set \( E \) in the sense of Vitali, that is, given an \( \epsilon > 0 \) and \( x \in E \), there is \( {aB} \in \mathcal{F} \) such that the diameter of \( B \) is less than \( \epsilon \) and \( x \in B \) ...
Proof. First we suppose \( E \) is bounded and contained in the bounded open set \( G \) . We disregard any members of \( \mathcal{F} \) that aren’t contained in \( G \) as well as those that don’t intersect \( E \) . The resulting family, which we will still refer to as \( \mathcal{F} \) covers \( E \) in the sense of...
No
Theorem 1.10. Any uncountable \( {\mathcal{G}}_{\delta } \) subset of \( \mathbb{R} \) contains a homeomorphic copy of the Cantor set \( \Delta \) .
Proof. Let \( E \) be such a set. Then there exists a descending sequence \( \left( {G}_{n}\right) \) of open sets in \( \mathbb{R} \) so that \( E = \mathop{\bigcap }\limits_{n}{G}_{n} \) . Let \( F \) be the set of all \( x \in E \) for which given any open set \( U \) containing \( x, U \cap E \) is uncountable. Not...
Yes
Theorem 1.11 (F. Bernstein). There exists a set \( B \) of real numbers such that both \( B \) and its complement \( {B}^{c} \) meet every uncountable closed set in \( \mathbb{R} \) .
Proof. We call on the well-ordering principle. Well-order \( \mathbb{R} \) . Using facts established above, we can also well-order the collection \( \mathcal{F} \) of all uncountable closed subsets of \( \mathbb{R} \) (each of which, is, by the way, of cardinality \( c \) ); we can index this well-ordering by the ordin...
Yes
Theorem 1.12 (Isoperimetric Inequality). Among convex bodies of a given volume, Euclidean balls have the least surface area.
Proof. Let \( C \) be a convex, compact set in \( {\mathbb{R}}^{n} \) whose \( n \) -dimensional volume is that of\n\n\[ B \mathrel{\text{:=}} {B}_{{l}_{n}^{2}}.\]\n\nThe surface area of \( C \) can be expressed by\n\n(1)\n\n\[ \operatorname{sa}\left( {\partial C}\right) = \mathop{\lim }\limits_{{\epsilon \searrow 0}}\...
Yes
Theorem 2.2 (Carathéodory). Let \( \mu \) be an outer measure on the set \( \Omega \) . Then\n\n(i) if \( \mu \left( N\right) = 0 \), then \( N \) is \( \mu \) -measurable;\n\n(ii) \( E \) is \( \mu \) -measurable if and only if \( {E}^{c} \) is;\n\n(iii) if \( \left( {E}_{n}\right) \) is a sequence of \( \mu \) -measu...
Proof. (i) Suppose \( A \subseteq N \) and \( B \subseteq {N}^{c} \) . Of course \( \mu \left( A\right) = 0 \) . Hence\n\n\[ \mu \left( B\right) \leq \mu \left( {A \cup B}\right) \leq \mu \left( A\right) + \mu \left( B\right) = \mu \left( B\right) .\n\nSqueezy says all are equal and so \( N \) is \( \mu \) -measurable....
Yes
Theorem 2.4. If \( \tau \) is a premeasure defined on a class \( \mathcal{C} \) of subsets of the space \( \Omega \), then the set function \( \mu \)\n\n\[ \mu \left( E\right) \mathrel{\text{:=}} \inf \left\{ {\mathop{\sum }\limits_{i}\tau \left( {C}_{i}\right) : {C}_{i} \in \mathcal{C}, E \subseteq \mathop{\bigcup }\l...
Proof. It is plain and easy to see that the set function \( \mu \) as defined from \( \tau \) satisfies properties (i)-(iii) of the definition of a measure. Only \
No
Theorem 2.5. \( \mu \) constructed by Method II is an outer measure.
Proof. The only possible stumbling point to this is the countable subadditivity, so let’s see why \( \mu \) is countably subadditive. To this end, let \( \left( {E}_{n}\right) \) be a sequence of subsets of \( \Omega \) and consider the quantities\n\n\[ \mu \left( {\mathop{\bigcup }\limits_{n}{E}_{n}}\right) \text{ and...
Yes
Theorem 2.8. If \( \mu \) is a metric outer measure on the metric space \( \Omega \), then every closed subset of \( \Omega \) is \( \mu \) -measurable. Consequently if \( \mu \) is a metric outer measure on the metric space \( \Omega \) , then every Borel subset of \( \Omega \) is \( \mu \) -measurable.
Proof. Suppose \( F \) is a closed subset of the metric space \( \Omega \) with metric \( \rho \) , and let \( A \subseteq F \) and \( B \subseteq {F}^{c} \) be nonempty sets. For each \( n \) let\n\n\[ \n{B}_{n} \mathrel{\text{:=}} \left\{ {x \in B : \mathop{\inf }\limits_{{y \in F}}\rho \left( {x, y}\right) > \frac{1...
Yes
Theorem 2.12. Let\n\n\[ \n{C}_{0} = \{ x = \left( {{x}_{1},\ldots ,{x}_{n}}\right) \subseteq {\mathbb{R}}^{n} : 0 \leq {x}_{i} < 1, i = 1,2,\ldots, n\} \n\] \n\nand \n\n\[ \n{\kappa }_{n} = {\mu }^{\left( n\right) }\left( {C}_{0}\right) \n\] \n\nThen \n\n\[ \n{\mu }^{\left( n\right) }\left( E\right) = {\kappa }_{n}{\la...
Proof. By definition, \n\n\[ \n{\mu }^{\left( n\right) }\left( {C}_{0}\right) = {\kappa }_{n}{\lambda }_{n}\left( {C}_{0}\right) \n\] \n\nSince both \( {\mu }^{\left( n\right) } \) and \( {\lambda }_{n} \) are invariant under translations and are homogeneous of order \( n \) (i.e., for any \( t \geq 0,{\mu }^{\left( n\...
Yes
Theorem 2.13. \[ {\kappa }_{n} = {\left( \frac{4}{\pi }\right) }^{n/2}\Gamma \left( \frac{n + 2}{2}\right) \]
Proof. To compute \( {\kappa }_{n} \) we need to know the volume, \( {\lambda }_{n}\left( {B}_{{\ell }_{2}^{n}}\right) \), of the closed unit ball in \( {\mathbb{R}}^{n} \) . Cutting to the quick, this is given by \[ {\lambda }_{n}\left( {B}_{{\ell }_{2}^{n}}\right) = \frac{{\pi }^{n/2}}{\Gamma \left( \frac{n + 2}{2}\r...
Yes
Theorem 3.1. Let \( G \) be a topological group. If \( U \) is an open set containing the identity \( e \), then there is an open set \( V \) containing \( e \) such that \( e \in V \subseteq \) \( \overline{V} \subseteq U. \) Consequently, a \( {T}_{0} \) topological group is regular and so Hausdorff.
Proof. First things first: Let \( U \) be an open set that contains the identity \( e \) . By continuity of multiplication, there is an open set \( W \) containing \( e \) such that \( W \cdot W \subseteq U \) . If we set \( V = W \cap {W}^{-1} \), then we have an open set that contains \( e \), is symmetric \( \left( ...
Yes
Proposition 3.2. Every open subgroup of a topological group is closed.
Proof. Let \( H \) be an open subgroup of the topological group \( G \) . Take \( g \in \) \( \bar{H} \) . Every open set that contains \( g \) intersects \( H;{gH} \) is such an open set.\n\nTherefore \( {gH} \cap H \neq \varnothing \) . Since cosets are either the same or disjoint, \( {gH} = H \) . Thus\n\n\[ g = {ge...
Yes
Corollary 3.7. \( {GL}\left( {n,\mathbb{C}}\right) \) is a locally compact metrizable topological group.
Proof. After all, \[ {GL}\left( {n,\mathbb{C}}\right) = \mathop{\det }\limits^{ \leftarrow }\left( {\{ z \in \mathbb{C} : z \neq 0\} }\right) \] and so \( {GL}\left( {n,\mathbb{C}}\right) \) is homeomorphic to an open subset of a locally compact metric space, \( {\mathbb{C}}^{{n}^{2}} \) . Our comments about continuity...
Yes
Theorem 3.8. The groups \( U\left( n\right), O\left( n\right) ,{SU}\left( n\right) \) , and \( {SO}\left( n\right) \) are compact metric topological groups.
Proof. Each of \( O\left( n\right) ,{SU}\left( n\right) ,{SO}\left( n\right) \) are closed subgroups of \( U\left( n\right) \), so it’s enough to establish that \( U\left( n\right) \) is compact. Now \( \alpha \in U\left( n\right) \) precisely when \( {\alpha }^{t}\bar{\alpha } = {\operatorname{id}}_{{\mathbb{C}}^{n}} ...
Yes
Corollary 3.10. Let \( G \) be a topological group. If \( G \) has a countable neighborhood base at \( \{ e\} \), then \( G \) is metrizable. In this case the metric can be taken to be left invariant.
Proof. Suppose \( \left\{ {{V}_{n} : n \in \mathbb{N}}\right\} \) is a countable open base at \( e \) . Let \( {U}_{1} = \) \( {V}_{1} \cap {V}_{1}^{-1} \), and let \( {U}_{2} \) be a symmetric open neighborhood of \( e \) such that \( {U}_{2} \subseteq {U}_{1} \cap {V}_{2} \) and \( {U}_{2} \cdot {U}_{2} \subseteq {U}...
Yes
Corollary 3.11. Let \( G \) be a topological group, let \( a \in G \), and let \( F \) be a closed subset of \( G \) such that \( a \notin F \) . Then there is a continuous real function \( \chi \) on \( G \) such that \( \chi \left( a\right) = 0 \) and \( \chi \left( x\right) = 1 \) for all \( x \in F \) . Consequentl...
Proof. Let \( {U}_{1} \) be a symmetric neighborhood of \( e \) such that \( \left( {a{U}_{1}}\right) \cap F = \varnothing \) . Choose a sequence \( \left( {{U}_{n} : n \geq 2}\right) \) of open neighborhoods of \( e \) such that each \( {U}_{n} \) is symmetric, \( {U}_{n + 1} \cdot {U}_{n + 1} \subseteq {U}_{n} \), an...
Yes
Theorem 3.12. Any locally compact topological group is paracompact, hence normal.
Proof. Let \( G \) be a locally compact topological group, and let \( V \) be an open set in \( G \) containing the identity of \( G \) and having a compact closure \( \bar{V} \) . By Theorem 3.1 there exists a symmetric open set \( U \) such that\n\n\[ e \in U \subseteq \bar{U} \subseteq U \cdot U \subseteq V. \]\n\nL...
Yes
Theorem 3.16. The product of Hausdorff spaces is a Hausdorff space.
Proof. If \( x, y \) are distinct members of \( S = \mathop{\prod }\limits_{\mathcal{A}}{S}_{\alpha } \), then there is an \( \alpha \in \mathcal{A} \) so that \( {x}_{\alpha } \neq {y}_{\alpha } \) ; since \( {S}_{\alpha } \) is a Hausdorff space there are open sets \( U \) and \( \begin{matrix} V & & \text{in}\;{S}_{...
No
Theorem 3.17 (Tychonoff’s Product Theorem). The product of compact spaces is compact.
Proof. (Chevalley and Frink [17]). Let \( \mathcal{F} \) be a family of subsets of the product \( S \) of the compact spaces \( {S}_{\alpha },\alpha \in \mathcal{A} \), and suppose that any finite subfamily of \( \mathcal{F} \) has a nonempty intersection. Our aim is to show\n\n\[ \mathop{\bigcap }\limits_{F}F \neq \va...
Yes
Theorem 4.1 (Hahn-Banach Theorem). Let \( X \) be a real linear space, and let \( S \) be a linear subspace of \( X \) . Suppose that \( p : X \rightarrow \mathbb{R} \) is a subadditive, positively homogeneous functional and \( f : S \rightarrow \mathbb{R} \) is a linear functional with \( f\left( s\right) \leq p\left(...
Proof. Our first task is to see how to extend a functional like \( f \) one dimension at a time while preserving the domination by \( p \) . With this in mind, let \( x \in X \smallsetminus S \), and notice that for any linear combination \( s + {\alpha x} \) of a vector in \( S \) and \( x \), whatever the linear exte...
No
Proposition 4.4. If \( G \) is a compact metrizable group, then \( A \cong B \) whenever \( A = {gB} \) for some \( g \in G \) is a congruence.
Proof. Let \( \left( {S}_{n}\right) \) be a sequence of concentric open balls (with common center \( c) \) with radii tending to zero. Suppose \( \left( {A}_{n}\right) \) satisfies \( {A}_{n} \cong {S}_{n} \) for each \( n \) and suppose \( {a}_{n},{b}_{n} \in {A}_{n} \) have \( \mathop{\lim }\limits_{n}{a}_{n} = a,\ma...
Yes
Proposition 4.5. Let \( Q \) be a compact metric space, and let \( G \) be a transitive equicontinuous group of homeomorphisms of \( Q \) onto itself. If \( A, B \in Q \), then \( A \cong B \) whenever \( A = g\left( B\right) \) for some \( g \in G \) is a congruence.
Proof. Transitivity of \( G \) is simply the feature that if \( {q}_{1},{q}_{2} \in Q \), then there is a \( g \in G \) so that \( g\left( {q}_{1}\right) = {q}_{2} \) . Because members of \( G \) are homeomorphisms of \( Q \) onto \( Q \) ,(iv) and (v) of Definition 4.3 are so. Transitivity of \( G \) assures us of (vi...
Yes
Proposition 4.6. Suppose \( A, B \), and \( C \) are nonempty open subsets of \( Q \) . Then\n\n(v) If \( d\left( {A, B}\right) = \) distance from \( A \) to \( B \) is positive (so \( \bar{A} \cap \bar{B} = \varnothing \) ) and \( \left( {S}_{n}\right) \) is sequence of open concentric balls with radii tending to zero...
Proof. Item (v) requires some serious and careful attention. Suppose (v) fails. Then there is \( \left( {n}_{k}\right) \) so that\n\n\[ h\left( {A \cup B,{S}_{{n}_{k}}}\right) < h\left( {A,{S}_{{n}_{k}}}\right) + h\left( {B,{S}_{{n}_{k}}}\right) \n\nfor each \( k \) . We can plainly suppose that the \( {n}_{k} \) ’s ar...
Yes
Proposition 4.7. If \( A \) and \( B \) are open sets, then\n\n(i) \( 0 < l\left( A\right) < \infty \), as long as \( A \neq \varnothing \).\n\n(ii) \( A \subseteq B \Rightarrow l\left( A\right) \leq l\left( B\right) \).\n\n(iii) \( l\left( {A \cup B}\right) \leq l\left( A\right) + l\left( B\right) \).\n\n(iv) \( A \co...
Proof. To see Proposition 4.7(v), note that by Proposition 4.6(v), there exists an \( N \) such that for all \( n \geq N \),\n\n\[ h\left( {A \cup B,{S}_{n}}\right) = h\left( {A,{S}_{n}}\right) + h\left( {B,{S}_{n}}\right) .\n\]\n\nFrom this we easily see that for all \( n \geq N \),\n\n\[ {l}_{n}\left( {A \cup B}\righ...
No
Theorem 4.8. Let \( Q \) be a compact metric space. Then\n\n(i) \( 0 \leq \lambda \left( X\right) \) .\n\n(ii) If \( X \) is a nonempty open subset of \( Q \), then \( 0 < \lambda \left( X\right) < \infty \) .\n\n(iii) \( X \subseteq Y \subseteq Q \Rightarrow \lambda \left( X\right) \leq \lambda \left( Y\right) \) .\n\...
Proof. Items (i)-(iv) tell us that \( \lambda \) is an outer measure, item (v) assures us that \( \lambda \) respects congruence, and item (vi) says that \( \lambda \) is a \
No
Lemma 4.11. For any \( x \in \mathcal{C},\bar{\int }\left( x\right) = f\left( x\right) \) .
Proof. On the one hand, we can let \( {x}_{n} = x \) for all \( n \) and \( m = x \) . This done, we plainly have\n\n\[ \mathop{\liminf }\limits_{n}{x}_{n} \geq x,\text{ and }{x}_{n} \geq m \]\n\nhence\n\n\[ \bar{\int }\left( x\right) \leq \mathop{\liminf }\limits_{n}f\left( {x}_{n}\right) = f\left( x\right) \]\n\nOn t...
Yes
Lemma 4.12. If \( {z}_{1},{z}_{2} \in {\mathcal{L}}^{ * } \) with \( \bar{\int }\left( {z}_{1}\right) ,\bar{\int }\left( {z}_{2}\right) < \infty \), then \( \bar{\int }\left( {{z}_{1} + {z}_{2}}\right) \leq \) \( \bar{\int }\left( {z}_{1}\right) + \bar{\int }\left( {z}_{2}\right) .
Proof. Suppose \( {P}_{1},{P}_{2} \) are numbers such that\n\n\[ \int \left( {z}_{1}\right) < {P}_{1}\text{ and }\int \left( {z}_{2}\right) < {P}_{2} \]\n\nThere are sequences \( \left( {x}_{n}^{\left( 1\right) }\right) ,\left( {x}_{n}^{\left( 2\right) }\right) \subseteq \mathcal{C} \) and functions \( {m}_{1},{m}_{2} ...
Yes
Lemma 4.13. For any \( z \in {\mathcal{L}}^{ * } \) , \[ \int \left( z\right) \leq \int \left( z\right) \]
Proof. There is nothing to prove if \( \bar{\int }\left( z\right) = + \infty \) . If \( \bar{\int }\left( {-z}\right) = + \infty \), then \( \int \left( z\right) = - \bar{\int }\left( {-z}\right) \), so again there is nothing to prove. If \( \bar{\int }\left( z\right) ,\bar{\int }\left( {-z}\right) < \infty \) , then L...
Yes
Lemma 4.14. If \( z \in {\mathcal{L}}^{ * } \) and \( \bar{f}\left( z\right) < \infty \), then\n\n\[ \int \left( \frac{z + \left| z\right| }{2}\right) < \infty \]\n\nand\n\n\[ \bar{\int }\left( z\right) = \bar{\int }\left( \frac{z + \left| z\right| }{2}\right) + \bar{\int }\left( \frac{z - \left| z\right| }{2}\right) ....
Proof. Suppose \( \bar{\int }\left( z\right) < P < \infty \) . Find \( m \in \mathcal{C} \) and \( \left( {x}_{n}\right) \subseteq \mathcal{C} \) so that \( {x}_{n} \geq m \) for all \( n \), and \( \mathop{\liminf }\limits_{n}{x}_{n} \geq z,\mathop{\lim }\limits_{n}f\left( {x}_{n}\right) < P \) . Notice that if \( {x}...
Yes
Lemma 4.18. If \( z \in \mathcal{L} \), then \( \left| z\right| \in \mathcal{L} \), that is, \( \mathcal{L} \) is a vector lattice.
Proof. Since\n\n\[ \left| z\right| = \left( \frac{z + \left| z\right| }{2}\right) + \left( \frac{\left| z\right| - z}{2}\right) \]\n\nit’s enough (thanks to \( \mathcal{L} \) ’s linearity) to show that \( \frac{z + \left| z\right| }{2},\frac{z - \left| z\right| }{2} \) both belong to \( \mathcal{L} \) if \( z \in \math...
Yes
Lemma 4.20 (Dominated Convergence Theorem). Suppose \( \left( {z}_{n}\right) \subseteq \mathcal{L}, M \in \) \( \mathcal{L} \) and \( \left| {z}_{n}\right| \leq M \) . Then\n\n\[ g = \mathop{\liminf }\limits_{n}{z}_{n}, h = \mathop{\limsup }\limits_{n}{z}_{n} \in \mathcal{L} \]\n\nwith\n\n\[ \bar{\int }\left( g\right) ...
Proof. For each \( i \) and for each \( j \geq i \), write\n\n\[ {g}_{ij} = \min \left\{ {{z}_{i},{z}_{i + 1},\ldots ,{z}_{j}}\right\} \]\n\nThen the sequence \( {\left( {g}_{ij}\right) }_{j = i}^{\infty } \) is decreasing, each member belongs to \( \mathcal{L} \), and so the sequence \( {\left( M - {g}_{ij}\right) }_{...
Yes
Lemma 4.21. If \( z \in \mathcal{L}, z \geq 0 \) and \( \bar{\int }z = 0 \), then whenever the function \( x \) satisfies \( \left| x\right| \leq z \), we have that \( x \in \mathcal{L} \) and \( \bar{\int }x = 0 \) .
This is an immediate consequence of Lemma 4.15.
No
Theorem 4.25. \( \mathcal{L} \) contains all of the indicator functions on Borel sets.
Suppose \( \left( {B}_{n}\right) \) is a sequence of pairwise disjoint Borel subsets of \( Q \) . Then \( {\chi }_{{B}_{n}},{\chi }_{\mathop{\bigcup }\limits_{n}{B}_{n}} \in \mathcal{L} \) and\n\n\[{\chi }_{\mathop{\bigcup }\limits_{n}{B}_{n}} = \mathop{\sum }\limits_{n}{\chi }_{{B}_{n}}\]\n\nan appeal to \( \left( {\m...
Yes
Theorem 4.26 (Hausdorff’s Paradox). The unit sphere \( {S}^{2} \) of \( {\mathbb{R}}^{3} \) can be decomposed into the disjoint union\n\n\[ \n{S}^{2} = Q \cup R \cup S \cup T \n\]\n\nof four sets: \( Q \) is countable, \( R, S, T \) are pairwise congruent, and \( R \) is congruent to \( S \cup T \) .
Now suppose that \( \mu \) is a solution to the easy problem of measure in \( {\mathbb{R}}^{3} \) . Normalize \( \mu \) so \( \mu \left( {B}_{{\mathbb{R}}^{3}}\right) = 1 \) . With Hausdorff’s paradox in hand, write \( {B}_{0} = {B}_{{\mathbb{R}}^{3}} \smallsetminus \{ 0\} \) in the form\n\n\[ \n{B}_{0} = {Q}_{0} \cup ...
No
Theorem 4.32. Let \( \Phi \) be a hyperfunction not belonging to the linear space \( \Omega \) of hyperfunctions, and suppose that there exist \( {F}_{1},{F}_{2} \in \Omega \) such that\n\n\[ \n{F}_{1} > \Phi > {F}_{2} \n\]\n\nLet \( A : \Omega \rightarrow \mathbb{R} \) be a positive linear functional. Then there exist...
The hypothesis of Theorem 4.32 tells us that\n\n\[ \n\alpha = \inf \{ A\left( F\right) : F \in \Omega, F > \Phi \} \n\]\n\nis real. Now if \( \Psi \in \Omega \left( \Phi \right) \) is written in the form\n\n\[ \n\Psi = F + {c\Phi } \n\]\n\nwhere \( c \in \mathbb{R}, F \in \Omega \), then\n\n\[ \n\bar{A}\left( \Psi \rig...
Yes
Theorem 4.35. Suppose \( f \) is Riemann integrable. Then \( f \sim c \) where \( c = \) \( {\int }_{0}^{1}f\left( x\right) {dx} \)
Proof. Suppose \( f \) is Riemann integrable, and define \( {f}_{n} \) on \( C \) by\n\n\[ \n{f}_{n}\left( x\right) = \frac{1}{n}\mathop{\sum }\limits_{{k = 1}}^{n}f\left( {x + \frac{k}{n}}\right) = \frac{1}{n}\mathop{\sum }\limits_{{k = 1}}^{n}{f}_{\frac{k}{n}}\left( x\right) .\n\]\n\nThen \( \left( {f}_{n}\right) \) ...
Yes
Proposition 4.37. There exists a bounded Lebesgue integrable function \( \rho \) with \( {\int }_{0}^{1}\rho \left( x\right) {dx} = 0 \) such that if \( \phi \) is Riemann integrable and \( \phi \succ \rho \), then \( {\int }_{0}^{1}\phi \left( x\right) {dx} > 1 \) .
If \( N \) is a set of Lebesgue measure zero with \( {N}^{c} \) a set of the first Baire category, then \( \rho = {\chi }_{N} \), the indicator function of \( N \), is bounded, Lebesgue integrable, and has \( {\int }_{0}^{1}\rho \left( x\right) {dx} = 0 \) . This \( \rho \) fits the bill.
Yes
Theorem 5.1. Let \( G \) be a topological group, and let \( M \) be a nonempty compact subset of \( G \) . Then any continuous function \( f : G \rightarrow \mathbb{R} \) is left uniformly continuous on \( M \) ; i.e., given \( \epsilon > 0 \), there is an open set \( V \) containing the identity of \( G \) so that if ...
Proof. Let \( \epsilon > 0 \) be given. For each \( a \in M \), there is an open set \( {V}_{a} \) that contains the identity such that if \( x \in M \) and \( x \in a{V}_{a} \), then \( \left| {f\left( x\right) - f\left( a\right) }\right| \leq \frac{\epsilon }{2} \) . Since \( e \cdot e = e \), there is an open set \(...
Yes
Theorem 5.4 (Arzelá and Ascoli). A nonempty subset \( \mathcal{K} \) of \( C\left( S\right) \) is relatively norm compact if and only if \( \mathcal{K} \) is uniformly bounded and equicontinuous.
Proof. Assume \( \mathcal{K} \) is uniformly bounded and equicontinuous. By Fact 5.3 the pointwise closure, \( \overline{{\mathcal{K}}^{p}} \) of \( \mathcal{K} \) is also equicontinuous. But \( \mathcal{K} \) is bounded in each coordinate by its uniform bound, and so \( \overline{{\mathcal{K}}^{p}} \) is also bounded ...
Yes
Corollary 5.5. Let \( G \) be a topological group, and let \( M \) be a nonempty compact subset of \( G \) . Suppose \( \mathcal{K} \subseteq C\left( G\right) \) is equicontinuous on \( M \) . Then given \( \epsilon > 0 \) there is an open set \( V \) containing the identity of \( G \) so that if \( x, y \in M \) and \...
Proof. (It is useful to take a close look at the proof of Theorem 5.1 before looking at this proof.)\n\nLet \( \epsilon > 0 \) be given. For each \( a \in M \) there is an open set \( {V}_{a} \) that contains the identity \( e \) of \( G \) such that if \( x \in M \) and \( x \in a{V}_{a} \), then \( \left| {f\left( x\...
Yes
Lemma 5.8. If \( f \in C\left( G\right) \) is not constant, then there is an \( F \in \mathcal{F}\left( G\right) \) such that\n\n\[ \n{\mathrm{{OscRAve}}}_{F}f < \mathrm{{Osc}}f.\n\]
Proof. After all, \( f \) ’s not being constant ensures that there is an \( \alpha \) such that \( \min f < \alpha < \max f \) . Set\n\n\[ \nU = \left\lbrack {f < \alpha }\right\rbrack = \{ x \in G : f\left( x\right) < \alpha \} .\n\]\n\nSince \( \min f < \alpha ,\;U \) is a nonempty open set in \( G \) and \( G = \mat...
Yes
Lemma 5.9. Let \( f \in C\left( G\right) \), and define \( \mathcal{K} = \left\{ {{\operatorname{RAve}}_{F}f : F \in \mathcal{F}\left( G\right) }\right\} \) . Then \( \mathcal{K} \) is uniformly bounded, equicontinuous family in \( C\left( G\right) \) .
Proof. The key to this precious fact is that \( f \) is of course uniformly continuous. So given an \( \epsilon > 0 \), there is an open set \( V \) in \( G \) containing \( G \) ’s identity such that if \( x{y}^{-1} \in V \), then \( \left| {f\left( x\right) - f\left( y\right) }\right| \leq \epsilon \) . Notice that i...
Yes
Lemma 5.10. Let \( f \in C\left( G\right) \) and \( \mathcal{K} = \left\{ {{\operatorname{RAve}}_{F}f : F \in \mathcal{F}\left( G\right) }\right\} \) . Then\n\n\[ \mathop{\inf }\limits_{{g \in \mathcal{K}}}\operatorname{Osc}g = 0 \]
Proof. Let\n\n\[ s = \mathop{\inf }\limits_{{g \in \mathcal{K}}}\operatorname{Osc}g = \inf \left\{ {{\operatorname{OscRAve}}_{F}f : F \in \mathcal{F}\left( G\right) }\right\} .\n\nTherefore there exists \( \left( {F}_{n}\right) \) in \( \mathcal{F}\left( G\right) \) such that Osc \( \left( {\text{RAve}}_{{F}_{n}}\right...
Yes
Theorem 5.11. Every \( f \in C\\left( G\\right) \) has a right mean.
Proof. By the techniques used in Lemma 5.10, there is a constant function \( h \) (say \( h\\left( x\\right) \\equiv p \) ) and a sequence \( \\left( {F}_{n}\\right) \\subseteq \\mathcal{F}\\left( G\\right) \) such that\n\n\[ \n\\mathop{\\lim }\\limits_{n}{\\begin{Vmatrix}{\\operatorname{RAve}}_{{F}_{n}}f - h\\end{Vmat...
Yes
Theorem 5.12. Let \( f \in C\left( G\right) \) . Let \( p \) be a right mean of \( f \), and let \( q \) be a left mean of \( f \) . Then \( p = q \) .
Proof. Let \( \epsilon > 0 \) . Find \( A, B \in \mathcal{F}\left( G\right) \) so that\n\n\[ \left| \right| {\mathrm{{RAve}}}_{A}f - p{\left| \right| }_{\infty } \leq \frac{\epsilon }{2},\;\left| \right| {\mathrm{{LAve}}}_{B}f - q{\left| \right| }_{\infty } \leq \frac{\epsilon }{2}. \]\n\nNow\n\n\[ {\operatorname{RAve}...
Yes
Theorem 5.15 (L. Kantorovitch). Let \( {x}^{ * } \) be a positive linear functional on \( C\left( K\right) \) . Then there is a positive linear functional \( {y}^{ * } \) on \( {l}^{\infty }\left( K\right) \) which is an extension of \( {x}^{ * } \) without increasing its norm.
The proof relies on the fact that \( C{\left( K\right) }^{ * } \) \
No
Theorem 5.17 (Alexandroff). Let \( \mu \) be a regular finitely additive nonnegative measure defined on the Borel field \( \mathcal{A} \) (the field generated by the open sets) of the compact Hausdorff space \( K \) . Then \( \mu \) is countably additive on \( \mathcal{A} \) .
Proof. Let \( \epsilon > 0 \) . Suppose \( \left( {E}_{n}\right) \) is a sequence of pairwise disjoint members of \( \mathcal{A} \) with \( E = \mathop{\bigcup }\limits_{n}{E}_{n} \in \mathcal{A} \) . There is an \( F \in \mathcal{A}, F \) closed, \( F \subseteq E \), with \( \mu \left( {E \smallsetminus F}\right) < \e...
Yes
Theorem 5.18. With bigamy an anathema, for each boy in \( B \) to be able to marry a girl with whom he is acquainted it is both necessary and sufficient that regardless of \( C \subseteq B \)\n\n\[ \left| {\mathop{\bigcup }\limits_{{b \in C}}G\left( b\right) }\right| \geq \left| C\right| \] holds.
Proof. Since it is plain that (25) is necessary, we'll concentrate on proving sufficiency of (25). So we suppose (25) to be in effect and prove the possibility of a wise matchmaker by an induction on the number of boys in \( B \) . Though not universally associated with marriage, we nevertheless introduce the notion of...
No
Lemma 5.20. If \( C \) and \( {C}^{\prime } \) are perfect, then so is \( C \cup {C}^{\prime } \) .
Proof. By (25) we know that\n\n\[ \left| {\mathop{\bigcup }\limits_{{b \in C \cup {C}^{\prime }}}G\left( b\right) }\right| \geq \left| {C \cup {C}^{\prime }}\right| \]\n\nFurther\n\n\[ \left| {C \cup {C}^{\prime }}\right| = \left| C\right| + \left| {C}^{\prime }\right| - \left| {C \cap {C}^{\prime }}\right| \]\n\nand, ...
Yes
Theorem 5.21 (Markov-Kakutani Fixed Point Theorem). Suppose \( K \) is a compact convex subset of the locally convex (Hausdorff) linear topological space \( E \), and suppose that \( T \) is a continuous linear operator \( T : E \rightarrow E \) such that \( T\left( K\right) \subseteq K \) . Then there is an element \(...
Proof. Fix \( {k}_{0} \in K \), and set\n\n\[ {k}_{n} = \frac{1}{n + 1}\left( {{k}_{0} + T\left( {k}_{0}\right) + \cdots + {T}^{n}\left( {k}_{0}\right) }\right) . \]\n\nEach \( {k}_{n} \in K \) . Because \( K \) is compact the sequence \( \left( {k}_{n}\right) \) has a limit point \( k \) . Let \( \phi \in {E}^{ * } \)...
Yes
Corollary 5.22. Suppose \( \mathcal{A} \) is a commuting family of continuous linear operators on the locally convex (Hausdorff) linear topological space, \( K \) is a compact convex subset of \( E \), and \( T\left( K\right) \subseteq K \) for each \( T \in \mathcal{A} \). Then there is \( {ak} \in K \) such that \[ T...
Proof. First we notice that if \( {T}_{1},\ldots ,{T}_{n} \in \mathcal{A} \), then there is a \( k \in K \) so that \[ {T}_{1}\left( k\right) = k,{T}_{2}\left( k\right) = k,\ldots ,{T}_{n}\left( k\right) = k. \] This follows by induction on \( n \). The Markov-Kakutani theorem shows it so for \( n = 1 \). Suppose we kn...
Yes
Corollary 5.25. Let \( G \) be a compact topological group. Then there exists a regular Borel probability \( \mu \) on \( G \) such that\n\n\[ \mu \left( {{g}^{-1}A}\right) = \mu \left( A\right) \]\n\nfor each \( g \in G \) and each Borel set \( A \subseteq G \) .
Proof. The proof proceeds along the same path as that outlined in the case when \( G \) is abelian. This time we see that the collection \( \left\{ {{T}_{g} : g \in G}\right\} \) is a group of isometries of \( C\left( G\right) \) onto \( C\left( G\right) \) and, as before,\n\n\[ {T}_{g}^{ * }\left( \mu \right) = {\mu }...
Yes
Theorem 6.1 (Weil). Let the compact group \( G \) act transitively on the compact Hausdorff space \( K \) . Then there is a closed subgroup \( H \) of \( G \) such that \( K \) and \( G/H \) are isomorphic under \( G \)’s action.
Proof. Fix \( {k}_{0} \in K \) . Look at\n\n\[ H = \left\{ {g \in G : g\left( {k}_{0}\right) = {k}_{0}}\right\} \]\n\n\( H \) is called an isotopy subgroup. It is plain that \( H \) is a closed subgroup of \( G \) . A natural candidate for the isomorphism of \( G/H \) and \( K \) is at hand: \( \phi : G/H \rightarrow K...
Yes
Theorem 6.2 (Weil). Suppose the compact group \( G \) acts transitively on the compact Hausdorff space \( K \) . Then there is a unique \( G \) -invariant regular Borel probability measure on \( K \) .
Proof. We identify \( K \) with \( G/H \) where \( H \) is an isotopy subgroup of \( G \) as in our previous theorem. Let\n\n\[ \n{q}_{H} : G \rightarrow G/H \n\] \n\nbe the natural quotient map. Suppose \( \mu \) is the normalized Haar measure on \( G \) and define \( {\mu }_{G/H} \) on \( G/H \) by \n\n\[ \n{\mu }_{G...
Yes
Lemma 6.8 (Schur’s Lemma). Let \( E \) and \( F \) be finite-dimensional complex linear spaces, and suppose \( \mathcal{R} \) and \( \mathcal{S} \) are irreducible families of linear transformations on \( E \) and \( F \), respectively. Assume that \( T : E \rightarrow F \) is a linear transformation such that given an...
Proof. Step One. Suppose \( \dim E = m,\dim F = n \), and \( \left\{ {{e}_{1},\ldots ,{e}_{m}}\right\} \) , \( \left\{ {{f}_{1},\ldots ,{f}_{n}}\right\} \) are a bases for \( E \) and \( F \), respectively. Let\n\n\[ \n{b}_{j} = T{e}_{j},\text{ and }L = \operatorname{span}\left\{ {{b}_{1},\ldots ,{b}_{m}}\right\} = T\l...
Yes
Corollary 6.9. Let \( E \) be a finite-dimensional complex linear space, and let \( \mathcal{R} \) be an irreducible family of linear transformations on \( E \) . Suppose \( A \) is a linear transformation on \( E \) such that\n\n\[ \n{AB} = {BA} \n\]\n\nfor each \( B \in \mathcal{R} \) . Then \( A = \alpha {\operatorn...
Proof. If \( A \neq 0 \) and \( \alpha \) is a nonzero eigenvalue of \( A \), then \( A - \alpha {\operatorname{id}}_{E} \) commutes with each \( B \in \mathcal{R} \) . Schur’s Lemma tells us either \( A - \alpha {\operatorname{id}}_{E} = 0 \) or \( A - \alpha {\operatorname{id}}_{E} \) is invertible. The latter is not...
Yes
Theorem 6.11. Let \( G \) be a compact topological group with normalized Haar measure \( \mu \) . Suppose \( x \rightarrow {U}_{x} = \left( {{u}_{{i}_{j}}\left( x\right) }\right) \) and \( x \rightarrow {V}_{x} = \left( {{v}_{kl}\left( x\right) }\right) \) are nonequivalent irreducible unitary representations of \( G \...
\[ {\int }_{G}{u}_{{i}_{j}}\left( x\right) \overline{{v}_{kl}\left( x\right) }{d\mu }\left( x\right) = 0 \] Proof. Let \( T : F \rightarrow E \) be a linear transformation. For any \( x \in G \), let \[ {A}_{x} = {U}_{x}T{V}_{x}^{-1} = {U}_{x}T{V}_{{x}^{-1}} = {U}_{x}T{V}_{x}^{ * }. \] Each \( {A}_{x} \) is a linear tr...
Yes
Theorem 6.12. The net \( {\left( {\kappa }_{U} * f\right) }_{U} \) converges to \( f \) in \( C\left( G\right) \) for each \( f \) .
Proof. Let \( f \in C\left( G\right) \) and \( \epsilon > 0 \) be given.\n\nBy \( f \) ’s uniform continuity, there is an open set \( W \) containing \( e \) such that regardless of \( s \in G \) if \( t \in W \), then\n\n\[ \left| {f\left( {{t}^{-1}s}\right) - f\left( s\right) }\right| \leq \epsilon \]\n\nLet \( U \) ...
No
Theorem 6.13 (Pietsch). Let \( u : X \rightarrow Y \) be a bounded linear operator between Banach spaces \( X \) and \( Y \) . Let \( K \subseteq {B}_{{X}^{ * }} \) be a weak*-compact, norming set. Then \( u \) is an absolutely p-summing operator if and only if there exists a \( \textit{regular Borel probability measur...
Pietsch's original proof was somewhat complicated, but in the early seventies a shorter proof was offered by B. Maurey [81]. We present an outline of Maurey's proof of the existence of the \
No
Corollary 7.8. If \( {f}_{1},{f}_{2} \in {\mathcal{L}}^{ + } \), then \( \bar{I}\left( {{f}_{1} + {f}_{2}}\right) = \bar{I}\left( {f}_{1}\right) + \bar{I}\left( {f}_{2}\right) \) .
This is an easy consequence of \( \phi \) ’s additivity on \( {\mathcal{K}}^{ + }\left( \mathcal{S}\right) \), and the fact that\n\n\[ \n{f}_{1} + {f}_{2} = \sup \left\{ {{g}_{1} + {g}_{2} : {g}_{1},{g}_{2} \in {\mathcal{K}}^{ + },{g}_{1} \leq {f}_{1},{g}_{2} \leq {f}_{2}}\right\} \n\]
Yes
Corollary 7.9. For any nonempty \( \mathcal{D} \subseteq {\mathcal{L}}^{ + } \) ,
\[ \bar{I}\left( {\mathop{\sum }\limits_{{f \in \mathcal{D}}}f}\right) = \mathop{\sum }\limits_{{f \in \mathcal{D}}}\bar{I}\left( f\right) \]
Yes
Theorem 7.12. If \( \left( {f}_{n}\right) \) is an ascending sequence of functions with \( {f}_{n} \) : \( S \rightarrow \left\lbrack {0,\infty }\right\rbrack \), then \[ {\int }_{\begin{matrix} {\text{ sup }{f}_{n} = \sup } \\ n \end{matrix}}{f}_{n} \]
Proof. It's plain from Fact 7.11(iii) that \[ \int \mathop{\sup }\limits_{n}{f}_{n} \geq \mathop{\sup }\limits_{n}\int {f}_{n} \] so we will establish the reverse inequality, namely, \[ \mathop{\sup }\limits_{n}\int {f}_{n} \geq {\int }_{n}\mathop{\sup }\limits_{n}{f}_{n} \] If \( \mathop{\sup }\limits_{n}\bar{\int }{f...
Yes
Theorem 7.15. (i) If \( \mathcal{U} \) is a family of pairwise disjoint open sets, then \[ \mu \left( {\mathop{\bigcup }\limits_{\mathcal{U}}U}\right) = \mathop{\sum }\limits_{\mathcal{U}}\mu \left( U\right) \]
Proof. To see item (i): if \( U \) is open, then \( {\chi }_{U} \in {\mathcal{L}}^{ + } \) . Indeed, if \( {s}_{0} \in U \) , then \( 1 = {\chi }_{U}\left( {s}_{0}\right) > h \) also holds for any \( s \in U \) ; i.e., \( {\chi }_{U}\left( s\right) > h \) . If \( {s}_{0} \notin U \) , then \( {\chi }_{U}\left( {s}_{0}\...
No
Theorem 7.17. All Borel subsets of \( S \) are \( \mu \) -measurable.
Proof. Suppose \( V \) is an open subset of \( \mathcal{S} \). Let \( U \) be an open subset in \( \mathcal{S} \) with \( \mu \left( U\right) < \infty \), and let \( \epsilon > 0 \) be given. Use Theorem 7.15 to find an open subset \( O \) of \( \mathcal{S} \) so that\n\n\[ U \cap {V}^{c} \subseteq O \]\n\nand\n\n(31)\...
Yes