This is the set of notes I wrote up for the talk I gave at the student analysis seminar on September 14.

* * * * *

Recall that we adopt the notation of Agnew and Morse [AM38] and write [\mathscr{L},\rho,f,X] to denote the following:

  1. X is a real vector space;
  2. \rho is a sublinear functional on X;
  3. f is a linear functional defined on a subspace Y of X;
  4. f is dominated by \rho, viz., f(x) \leq \rho(x) for all x \in Y;
  5. \Lambda is a collection of endomorphisms on X;
  6. \rho is invariant under each \Lambda, viz., \rho ( \Lambda x ) = \rho(x) for each \Lambda \in \mathscr{L};
  7. f is invariant under each \Lambda, viz., f(\Lambda x) = f(x) for each \Lambda \in \mathscr{L};
  8. Y is invariant under each \Lambda, viz., \Lambda(Y) \subseteq Y for each \Lambda \in \mathscr{L}.

Furthermore, if [\mathscr{L},\rho,f,X] is satisfied, we write \{\mathscr{L},\rho,f,X\} to denote the collection of extensions F of f on X such that F is still dominated by \rho, and that

F ( \Lambda x ) = F(x)

for all \Lambda \in \mathscr{L}. In this notation, the classical Hahn-Banach theorem can be phrased as follows:

Theorem 1 (Hahn-Banach). Let X be a real vector space, and \mathscr{L}=\{I\}, where I:X \to X is the identity mapping. If [\mathscr{L},\rho,f,X], then \{\mathscr{L},\rho,f,X\} is nonempty.

Recall that a semigroup is a set with an associative binary operation. In the last talk, we have proved the following symmetry-invariant extension of Hahn-Banach:

Theorem 2 (Agnew-Morse). Let \mathscr{L} be a semigroup of commuting endomorphisms on a real vector space X. If [\mathscr{L},\rho,f,X], then \{\mathscr{L},\rho,f,X] is nonempty.

For a proof, see pages 25-27 of Lax [Lax02].

We also recall the key example from the last talk. We write X^* to denote the space of all real-valued sequences, and X_b to denote the space l^\infty(\mathbb{R}) of all bounded real-valued sequences. The classical Banach limit extend the usual limit in the following manner:

Theorem 3. There exists an operator L:X_b \to X_b, called a generalized limit or a Banach limit, such that

  1. The generalized limit agrees with the usual limit if the usual limit exists;
  2. L ((x_n)_{n=1}^\infty + (y_n)_{n=1}^\infty)  = L( (x_n)_{n=1}^\infty ) + L( (y_n)_{n=1}^\infty );
  3. L( (x_n)_{n=1}^\infty ) = L( (x_n)_{n=k}^\infty) for any k \in \mathbb{N};
  4. \displaystyle \liminf_{n \to \infty} x_n \leq L( (x_n)_{n=1}^\infty ) \leq \limsup_{n \to \infty} x_n.

We now see that the above theorem is an easy consequence of the Agnew-Morse theorem. Indeed, the Agnew-Morse theorem furnishes an extension of the regular limit operator that is invariant under the translation operator. A slight modification of the Agnew-Morse theorem allows us to extend the limit operator even further. First, some definitions:

Definition 4. X^* is the space of all real-valued sequences. Let x = (x_n)_{n=1}^\infty \in X^*.

  • T:X^* \to X^* is the translation operator

    T(x) = (x_2,x_3,\ldots).

  • H:X^* \to X^* is the Hölder mean operator

    H(x) = (x_1,[x_1+x_2]/2,[x_1+x_2+x_3]/3,\ldots).

  • For each integer r>0, the map I_r:H^* \to H^* is the r-fold iteration operator

    I_r(x) = (x_1,\ldots,x_1,x_2,\ldots,x_2,x_3\ldots,x_3,\ldots),

    where each term is repeated r times.

Definition 5. We define the following subspaces of $X^*$..

  • X_0 is the space of all convergent sequences with limit 0.
  • X_c is the space of all convergent sequences.
  • X_b= l^\infty(\mathbb{R}) is the space of all bounded sequences.
  • For each integer k>0, X_k is the space of all x \in X^* such that H^k x \in X_b.
  • X_{o(n)} is the set of all sequences (x_1,\ldots,x_n,\ldots) in X^* such that |x_n| = o(n), that is, |x_n| is dominated by n asymptotically.
  • X = X_{o(n)} \cap [\bigcup_{k=1}^\infty X_k].

The extended result is the following, which is Theorem 4 in [MWW69]:

Theorem 6. On \bigcup_{k=1}^\infty X_k, there exists a positive linear functional f_1 such that f_1(x) is the limit of the sequence H^k x whenever H^k x \in X_c. Furthermore, f_1 on \bigcup_{k=1}^\infty is invariant under all elements in the semigroup generated by H and T, and f_1 on X is invariant under all elements in the semigroup generated by H, T, and I_r, where r \in \mathbb{N}.

We omit the proof of this result. Interested readers can find a proof in [MWW69]. We merely point out the variant of the Agnew-Morse theorem required to establish the above theorem. To state the variant, we need some definitions.

Definition 7. A set G of endomorphisms on X is said to act on a subset X_1 of X commutatively to within left H-factors if each x \in X_1 and every g_1,g_2 \in G furnish h_1,h_2 \in H such that

h_1g_1g_2(x) = h_2g_2g_1(x).

Definition 8. Let G be s set of endomorphisms on X and X_1 a subset of X. We say that G is left-solvable over X_1 if there exists a sequence

G = G_n \supseteq G_{n-1} \supseteq \cdots \supseteq G_0

of endomorphisms such that, for each 0 \leq k leq n-1, the set G_{k+1} acts on X_1 commutatively to within left G_k-factors, and if G_0 is commutative.

Here is the required variant, which is Theorem 2 in [MWW69]; interested readers can find a proof in [MWW69]:

Theorem 9. Let X be a partially ordered linear space and G a set of order-preserving linear transformations of X into itself. Let

X_0 \subseteq X_1 \subseteq \cdots

be a set of \bar{n} G-invariant subspaces (\bar{n} may be infinity) such that the union is X. We assume that, for each 1 < n < \bar{n}, either

  1. G is left-solvable over X_n, and for each x \in X, there is an s \in X_{n-1} such that s \geq x; or
  2. for each x \in X_n, there is a g \in G such that g(x) \in X_{n-1}; and for each x \in X_n and g_1,g_2,g_3 \in G such that g_1(x) and g_2g_3(x) are in X_{n-1}, there are h_1,h_2 \in G such that

    h_1g_1g_2g_3(x) = h_2g_2g_3g_1(x).

Then every G-invariant positive linear functional on X_0 can be extended to a G-invariant positive linear functional on X.

As another application of the Agnew-Morse theorem, we consider generalizations of the Lebesgue measure:

Theorem 10. There exists a nonnegative finitely additive set function m: \mathcal{P}(\mathbb{S}^1) \to [0,\infty) that is invariant under rotation.

A proof was presented in the talk, which is taken directly from [Lax02]. Interested readers can find a proof in pages 33-34 of [Lax02]. Powered by Theorem 9, we can generalize the above theorem to the following:

Theorem 11. Let X be a compact Hausdorff space and G a semigroup of continuous endomorphisms on X containing the identity such that G is right-solvable over itself. Then there is a Baire measure on X which is G-invariant and ergodic.

This is Theorem 6 in [MWW69]. No proof of the above result was presented in the talk. The paper contains a partial proof.

References

  • [AM38] A. G. Agnew and A. P. Morse, “Extensions of linear functionals, with applications to limits, integrals, measures, and densities,” Annals of Mathematics 39 (1938), no. 1, 20-30.
  • [Lax02] Peter D. Lax, Functional Analysis, John Wiley & Sons. 2002.
  • [MWW69] E. J. McShane, R. B. Warfield, and V. M. Warfield. “Invariant extensions of linear functionals, with applications to measures and stochastic processes,” Pacific Journal of Mathematics 28 (1969), no. 1, 121-142.

 Leave a Reply

(required)

(required)

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

   
© 2011 Mark Hyun-ki Kim Suffusion theme by Sayontan Sinha