Misplaced Pages

In mathematical analysis, Lorentz spaces, introduced by George G. Lorentz in the 1950s, are generalisations of the more familiar ${\displaystyle L^{p}}$ spaces.

The Lorentz spaces are denoted by ${\displaystyle L^{p,q}}$. Like the ${\displaystyle L^{p}}$ spaces, they are characterized by a norm (technically a quasinorm) that encodes information about the "size" of a function, just as the ${\displaystyle L^{p}}$ norm does. The two basic qualitative notions of "size" of a function are: how tall is the graph of the function, and how spread out is it. The Lorentz norms provide tighter control over both qualities than the ${\displaystyle L^{p}}$ norms, by exponentially rescaling the measure in both the range (${\displaystyle p}$) and the domain (${\displaystyle q}$). The Lorentz norms, like the ${\displaystyle L^{p}}$ norms, are invariant under arbitrary rearrangements of the values of a function.

Definition

The Lorentz space on a measure space ${\displaystyle (X,\mu )}$ is the space of complex-valued measurable functions ${\displaystyle f}$ on X such that the following quasinorm is finite

${\displaystyle \|f\|_{L^{p,q}(X,\mu )}=p^{\frac {1}{q}}\left\|t\mu \{|f|\geq t\}^{\frac {1}{p}}\right\|_{L^{q}\left(\mathbf {R} ^{+},{\frac {dt}{t}}\right)}}$

where ${\displaystyle 0<p<\infty }$ and ${\displaystyle 0<q\leq \infty }$. Thus, when ${\displaystyle q<\infty }$,

${\displaystyle \|f\|_{L^{p,q}(X,\mu )}=p^{\frac {1}{q}}\left(\int _{0}^{\infty }t^{q}\mu \left\{x:|f(x)|\geq t\right\}^{\frac {q}{p}}\,{\frac {dt}{t}}\right)^{\frac {1}{q}}=\left(\int _{0}^{\infty }{\bigl (}\tau \mu \left\{x:|f(x)|^{p}\geq \tau \right\}{\bigr )}^{\frac {q}{p}}\,{\frac {d\tau }{\tau }}\right)^{\frac {1}{q}}.}$

and, when ${\displaystyle q=\infty }$,

${\displaystyle \|f\|_{L^{p,\infty }(X,\mu )}^{p}=\sup _{t>0}\left(t^{p}\mu \left\{x:|f(x)|>t\right\}\right).}$

It is also conventional to set ${\displaystyle L^{\infty ,\infty }(X,\mu )=L^{\infty }(X,\mu )}$.

Decreasing rearrangements

The quasinorm is invariant under rearranging the values of the function ${\displaystyle f}$, essentially by definition. In particular, given a complex-valued measurable function ${\displaystyle f}$ defined on a measure space, ${\displaystyle (X,\mu )}$, its decreasing rearrangement function, ${\displaystyle f^{\ast }:}$ can be defined as

${\displaystyle f^{\ast }(t)=\inf\{\alpha \in \mathbf {R} ^{+}:d_{f}(\alpha )\leq t\}}$

where ${\displaystyle d_{f}}$ is the so-called distribution function of ${\displaystyle f}$, given by

${\displaystyle d_{f}(\alpha )=\mu (\{x\in X:|f(x)|>\alpha \}).}$

Here, for notational convenience, ${\displaystyle \inf \varnothing }$ is defined to be ${\displaystyle \infty }$.

The two functions ${\displaystyle |f|}$ and ${\displaystyle f^{\ast }}$ are equimeasurable, meaning that

${\displaystyle \lambda {\bigl (}\{x\in X:|f(x)|>\alpha \}{\bigr )}=\lambda {\bigl (}\{t>0:f^{\ast }(t)>\alpha \}{\bigr )},\quad \alpha >0,}$

where ${\displaystyle \lambda }$ is the Lebesgue measure on the real line. The related symmetric decreasing rearrangement function, which is also equimeasurable with ${\displaystyle f}$, would be defined on the real line by

${\displaystyle \mathbf {R} \ni t\mapsto {\tfrac {1}{2}}f^{\ast }(|t|).}$

Given these definitions, for ${\displaystyle 0<p<\infty }$ and ${\displaystyle 0<q\leq \infty }$, the Lorentz quasinorms are given by

${\displaystyle \|f\|_{L^{p,q}}={\begin{cases}\left(\displaystyle \int _{0}^{\infty }\left(t^{\frac {1}{p}}f^{\ast }(t)\right)^{q}\,{\frac {dt}{t}}\right)^{\frac {1}{q}}&q\in (0,\infty ),\\\sup \limits _{t>0}\,t^{\frac {1}{p}}f^{\ast }(t)&q=\infty .\end{cases}}}$

Lorentz sequence spaces

When ${\displaystyle (X,\mu )=(\mathbb {N} ,\#)}$ (the counting measure on ${\displaystyle \mathbb {N} }$), the resulting Lorentz space is a sequence space. However, in this case it is convenient to use different notation.

Definition.

For ${\displaystyle (a_{n})_{n=1}^{\infty }\in \mathbb {R} ^{\mathbb {N} }}$ (or ${\displaystyle \mathbb {C} ^{\mathbb {N} }}$ in the complex case), let ${\textstyle \left\|(a_{n})_{n=1}^{\infty }\right\|_{p}=\left(\sum _{n=1}^{\infty }|a_{n}|^{p}\right)^{1/p}}$ denote the p-norm for ${\displaystyle 1\leq p<\infty }$ and ${\textstyle \left\|(a_{n})_{n=1}^{\infty }\right\|_{\infty }=\sup _{n\in \mathbb {N} }|a_{n}|}$ the ∞-norm. Denote by ${\displaystyle \ell _{p}}$ the Banach space of all sequences with finite p-norm. Let ${\displaystyle c_{0}}$ the Banach space of all sequences satisfying ${\displaystyle \lim _{n\to \infty }a_{n}=0}$, endowed with the ∞-norm. Denote by ${\displaystyle c_{00}}$ the normed space of all sequences with only finitely many nonzero entries. These spaces all play a role in the definition of the Lorentz sequence spaces ${\displaystyle d(w,p)}$ below.

Let ${\displaystyle w=(w_{n})_{n=1}^{\infty }\in c_{0}\setminus \ell _{1}}$ be a sequence of positive real numbers satisfying ${\displaystyle 1=w_{1}\geq w_{2}\geq w_{3}\geq \cdots }$, and define the norm ${\textstyle \left\|(a_{n})_{n=1}^{\infty }\right\|_{d(w,p)}=\sup _{\sigma \in \Pi }\left\|(a_{\sigma (n)}w_{n}^{1/p})_{n=1}^{\infty }\right\|_{p}}$. The Lorentz sequence space ${\displaystyle d(w,p)}$ is defined as the Banach space of all sequences where this norm is finite. Equivalently, we can define ${\displaystyle d(w,p)}$ as the completion of ${\displaystyle c_{00}}$ under ${\displaystyle \|\cdot \|_{d(w,p)}}$.

Properties

The Lorentz spaces are genuinely generalisations of the ${\displaystyle L^{p}}$ spaces in the sense that, for any ${\displaystyle p}$, ${\displaystyle L^{p,p}=L^{p}}$, which follows from Cavalieri's principle. Further, ${\displaystyle L^{p,\infty }}$ coincides with weak ${\displaystyle L^{p}}$. They are quasi-Banach spaces (that is, quasi-normed spaces which are also complete) and are normable for ${\displaystyle 1<p<\infty }$ and ${\displaystyle 1\leq q\leq \infty }$. When ${\displaystyle p=1}$, ${\displaystyle L^{1,1}=L^{1}}$ is equipped with a norm, but it is not possible to define a norm equivalent to the quasinorm of ${\displaystyle L^{1,\infty }}$, the weak ${\displaystyle L^{1}}$ space. As a concrete example that the triangle inequality fails in ${\displaystyle L^{1,\infty }}$, consider

${\displaystyle f(x)={\tfrac {1}{x}}\chi _{(0,1)}(x)\quad {\text{and}}\quad g(x)={\tfrac {1}{1-x}}\chi _{(0,1)}(x),}$

whose ${\displaystyle L^{1,\infty }}$ quasi-norm equals one, whereas the quasi-norm of their sum ${\displaystyle f+g}$ equals four.

The space ${\displaystyle L^{p,q}}$ is contained in ${\displaystyle L^{p,r}}$ whenever ${\displaystyle q<r}$. The Lorentz spaces are real interpolation spaces between ${\displaystyle L^{1}}$ and ${\displaystyle L^{\infty }}$.

Hölder's inequality

${\displaystyle \|fg\|_{L^{p,q}}\leq A_{p_{1},p_{2},q_{1},q_{2}}\|f\|_{L^{p_{1},q_{1}}}\|g\|_{L^{p_{2},q_{2}}}}$ where ${\displaystyle 0<p,p_{1},p_{2}<\infty }$, ${\displaystyle 0<q,q_{1},q_{2}\leq \infty }$, ${\displaystyle 1/p=1/p_{1}+1/p_{2}}$, and ${\displaystyle 1/q=1/q_{1}+1/q_{2}}$.

Dual space

If ${\displaystyle (X,\mu )}$ is a nonatomic σ-finite measure space, then
(i) ${\displaystyle (L^{p,q})^{*}=\{0\}}$ for ${\displaystyle 0<p<1}$, or ${\displaystyle 1=p<q<\infty }$;
(ii) ${\displaystyle (L^{p,q})^{*}=L^{p',q'}}$ for ${\displaystyle 1<p<\infty ,0<q\leq \infty }$, or ${\displaystyle 0<q\leq p=1}$;
(iii) ${\displaystyle (L^{p,\infty })^{*}\neq \{0\}}$ for ${\displaystyle 1\leq p\leq \infty }$.
Here ${\displaystyle p'=p/(p-1)}$ for ${\displaystyle 1<p<\infty }$, ${\displaystyle p'=\infty }$ for ${\displaystyle 0<p\leq 1}$, and ${\displaystyle \infty '=1}$.

Atomic decomposition

The following are equivalent for ${\displaystyle 0<p\leq \infty ,1\leq q\leq \infty }$.
(i) ${\displaystyle \|f\|_{L^{p,q}}\leq A_{p,q}C}$.
(ii) ${\displaystyle f=\textstyle \sum _{n\in \mathbb {Z} }f_{n}}$ where ${\displaystyle f_{n}}$ has disjoint support, with measure ${\displaystyle \leq 2^{n}}$, on which ${\displaystyle 0<H_{n+1}\leq |f_{n}|\leq H_{n}}$ almost everywhere, and ${\displaystyle \|H_{n}2^{n/p}\|_{\ell ^{q}(\mathbb {Z} )}\leq A_{p,q}C}$.
(iii) ${\displaystyle |f|\leq \textstyle \sum _{n\in \mathbb {Z} }H_{n}\chi _{E_{n}}}$ almost everywhere, where ${\displaystyle \mu (E_{n})\leq A_{p,q}'2^{n}}$ and ${\displaystyle \|H_{n}2^{n/p}\|_{\ell ^{q}(\mathbb {Z} )}\leq A_{p,q}C}$.
(iv) ${\displaystyle f=\textstyle \sum _{n\in \mathbb {Z} }f_{n}}$ where ${\displaystyle f_{n}}$ has disjoint support ${\displaystyle E_{n}}$, with nonzero measure, on which ${\displaystyle B_{0}2^{n}\leq |f_{n}|\leq B_{1}2^{n}}$ almost everywhere, ${\displaystyle B_{0},B_{1}}$ are positive constants, and ${\displaystyle \|2^{n}\mu (E_{n})^{1/p}\|_{\ell ^{q}(\mathbb {Z} )}\leq A_{p,q}C}$.
(v) ${\displaystyle |f|\leq \textstyle \sum _{n\in \mathbb {Z} }2^{n}\chi _{E_{n}}}$ almost everywhere, where ${\displaystyle \|2^{n}\mu (E_{n})^{1/p}\|_{\ell ^{q}(\mathbb {Z} )}\leq A_{p,q}C}$.

See also

• Interpolation space
• Hardy–Littlewood inequality

References

• Grafakos, Loukas (2008), Classical Fourier analysis, Graduate Texts in Mathematics, vol. 249 (2nd ed.), Berlin, New York: Springer-Verlag, doi:10.1007/978-0-387-09432-8, ISBN 978-0-387-09431-1, MR 2445437.

Notes

1. G. Lorentz, "Some new function spaces", Annals of Mathematics 51 (1950), pp. 37-55.
2. G. Lorentz, "On the theory of spaces Λ", Pacific Journal of Mathematics 1 (1951), pp. 411-429.

v t e Lp spaces
Basic concepts	Banach & Hilbert spaces L spaces Measure Lebesgue Measure space Measurable space/function Minkowski distance Sequence spaces
L spaces	Integrable function Lebesgue integration Taxicab geometry
L spaces	Bessel's Cauchy–Schwarz Euclidean distance Hilbert space Parseval's identity Polarization identity Pythagorean theorem Square-integrable function
${\displaystyle L^{\infty }}$ spaces	Bounded function Chebyshev distance Infimum and supremum Essential Uniform norm
Maps	Almost everywhere Convergence almost everywhere Convergence in measure Function space Integral transform Locally integrable function Measurable function Symmetric decreasing rearrangement
Inequalities	Babenko–Beckner Chebyshev's Clarkson's Hanner's Hausdorff–Young Hölder's Markov's Minkowski Young's convolution
Results	Marcinkiewicz interpolation theorem Plancherel theorem Riemann–Lebesgue Riesz–Fischer theorem Riesz–Thorin theorem For Lebesgue measure Isoperimetric inequality Brunn–Minkowski theorem Milman's reverse Minkowski–Steiner formula Prékopa–Leindler inequality Vitale's random Brunn–Minkowski inequality
Applications & related	Bochner space Fourier analysis Lorentz space Probability theory Quasinorm Real analysis Sobolev space *-algebra C*-algebra Von Neumann

• Banach spaces
• Lp spaces