{"title":"Higher order Lipschitz Sandwich theorems","authors":"Terry Lyons, Andrew D. McLeod","doi":"10.1112/jlms.70121","DOIUrl":null,"url":null,"abstract":"<p>We investigate the consequence of two <span></span><math>\n <semantics>\n <mrow>\n <mi>Lip</mi>\n <mo>(</mo>\n <mi>γ</mi>\n <mo>)</mo>\n </mrow>\n <annotation>${\\mathrm{Lip}}(\\gamma)$</annotation>\n </semantics></math> functions, in the sense of Stein, being close throughout a subset of their domain. A particular consequence of our results is the following. Given <span></span><math>\n <semantics>\n <mrow>\n <msub>\n <mi>K</mi>\n <mn>0</mn>\n </msub>\n <mo>></mo>\n <mi>ε</mi>\n <mo>></mo>\n <mn>0</mn>\n </mrow>\n <annotation>$K_0 > \\varepsilon > 0$</annotation>\n </semantics></math> and <span></span><math>\n <semantics>\n <mrow>\n <mi>γ</mi>\n <mo>></mo>\n <mi>η</mi>\n <mo>></mo>\n <mn>0</mn>\n </mrow>\n <annotation>$\\gamma > \\eta > 0$</annotation>\n </semantics></math>, there is a constant <span></span><math>\n <semantics>\n <mrow>\n <mi>δ</mi>\n <mo>=</mo>\n <mi>δ</mi>\n <mo>(</mo>\n <mi>γ</mi>\n <mo>,</mo>\n <mi>η</mi>\n <mo>,</mo>\n <mi>ε</mi>\n <mo>,</mo>\n <msub>\n <mi>K</mi>\n <mn>0</mn>\n </msub>\n <mo>)</mo>\n <mo>></mo>\n <mn>0</mn>\n </mrow>\n <annotation>$\\delta = \\delta (\\gamma,\\eta,\\varepsilon,K_0) > 0$</annotation>\n </semantics></math> for which the following is true. Let <span></span><math>\n <semantics>\n <mrow>\n <mi>Σ</mi>\n <mo>⊂</mo>\n <msup>\n <mi>R</mi>\n <mi>d</mi>\n </msup>\n </mrow>\n <annotation>$\\Sigma \\subset {\\mathbb {R}}^d$</annotation>\n </semantics></math> be closed and <span></span><math>\n <semantics>\n <mrow>\n <mi>f</mi>\n <mo>,</mo>\n <mi>h</mi>\n <mo>:</mo>\n <mi>Σ</mi>\n <mo>→</mo>\n <mi>R</mi>\n </mrow>\n <annotation>$f, h: \\Sigma \\rightarrow {\\mathbb {R}}$</annotation>\n </semantics></math> be <span></span><math>\n <semantics>\n <mrow>\n <mi>Lip</mi>\n <mo>(</mo>\n <mi>γ</mi>\n <mo>)</mo>\n </mrow>\n <annotation>${\\mathrm{Lip}}(\\gamma)$</annotation>\n </semantics></math> functions whose <span></span><math>\n <semantics>\n <mrow>\n <mi>Lip</mi>\n <mo>(</mo>\n <mi>γ</mi>\n <mo>)</mo>\n </mrow>\n <annotation>${\\mathrm{Lip}}(\\gamma)$</annotation>\n </semantics></math> norms are both bounded above by <span></span><math>\n <semantics>\n <msub>\n <mi>K</mi>\n <mn>0</mn>\n </msub>\n <annotation>$K_0$</annotation>\n </semantics></math>. Suppose <span></span><math>\n <semantics>\n <mrow>\n <mi>B</mi>\n <mo>⊂</mo>\n <mi>Σ</mi>\n </mrow>\n <annotation>$B \\subset \\Sigma$</annotation>\n </semantics></math> is closed and that <span></span><math>\n <semantics>\n <mi>f</mi>\n <annotation>$f$</annotation>\n </semantics></math> and <span></span><math>\n <semantics>\n <mi>h</mi>\n <annotation>$h$</annotation>\n </semantics></math> coincide throughout <span></span><math>\n <semantics>\n <mi>B</mi>\n <annotation>$B$</annotation>\n </semantics></math>. Then, over the set of points in <span></span><math>\n <semantics>\n <mi>Σ</mi>\n <annotation>$\\Sigma$</annotation>\n </semantics></math> whose distance to <span></span><math>\n <semantics>\n <mi>B</mi>\n <annotation>$B$</annotation>\n </semantics></math> is at most <span></span><math>\n <semantics>\n <mi>δ</mi>\n <annotation>$\\delta$</annotation>\n </semantics></math>, we have that the <span></span><math>\n <semantics>\n <mrow>\n <mi>Lip</mi>\n <mo>(</mo>\n <mi>η</mi>\n <mo>)</mo>\n </mrow>\n <annotation>${\\mathrm{Lip}}(\\eta)$</annotation>\n </semantics></math> norm of the difference <span></span><math>\n <semantics>\n <mrow>\n <mi>f</mi>\n <mo>−</mo>\n <mi>h</mi>\n </mrow>\n <annotation>$f-h$</annotation>\n </semantics></math> is bounded above by <span></span><math>\n <semantics>\n <mi>ε</mi>\n <annotation>$\\varepsilon$</annotation>\n </semantics></math>. More generally, we establish that this phenomenon remains valid in a less restrictive Banach space setting under the weaker hypothesis that the two <span></span><math>\n <semantics>\n <mrow>\n <mi>Lip</mi>\n <mo>(</mo>\n <mi>γ</mi>\n <mo>)</mo>\n </mrow>\n <annotation>${\\mathrm{Lip}}(\\gamma)$</annotation>\n </semantics></math> functions <span></span><math>\n <semantics>\n <mi>f</mi>\n <annotation>$f$</annotation>\n </semantics></math> and <span></span><math>\n <semantics>\n <mi>h</mi>\n <annotation>$h$</annotation>\n </semantics></math> are only close in a pointwise sense throughout the closed subset <span></span><math>\n <semantics>\n <mi>B</mi>\n <annotation>$B$</annotation>\n </semantics></math>. We require only that the subset <span></span><math>\n <semantics>\n <mi>Σ</mi>\n <annotation>$\\Sigma$</annotation>\n </semantics></math> be closed; in particular, the case that <span></span><math>\n <semantics>\n <mi>Σ</mi>\n <annotation>$\\Sigma$</annotation>\n </semantics></math> is finite is covered by our results. The restriction that <span></span><math>\n <semantics>\n <mrow>\n <mi>η</mi>\n <mo><</mo>\n <mi>γ</mi>\n </mrow>\n <annotation>$\\eta < \\gamma$</annotation>\n </semantics></math> is sharp in the sense that our result is false for <span></span><math>\n <semantics>\n <mrow>\n <mi>η</mi>\n <mo>:</mo>\n <mo>=</mo>\n <mi>γ</mi>\n </mrow>\n <annotation>$\\eta:= \\gamma$</annotation>\n </semantics></math>.</p>","PeriodicalId":49989,"journal":{"name":"Journal of the London Mathematical Society-Second Series","volume":"111 3","pages":""},"PeriodicalIF":1.0000,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1112/jlms.70121","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of the London Mathematical Society-Second Series","FirstCategoryId":"100","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1112/jlms.70121","RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATHEMATICS","Score":null,"Total":0}
引用次数: 0
Abstract
We investigate the consequence of two functions, in the sense of Stein, being close throughout a subset of their domain. A particular consequence of our results is the following. Given and , there is a constant for which the following is true. Let be closed and be functions whose norms are both bounded above by . Suppose is closed and that and coincide throughout . Then, over the set of points in whose distance to is at most , we have that the norm of the difference is bounded above by . More generally, we establish that this phenomenon remains valid in a less restrictive Banach space setting under the weaker hypothesis that the two functions and are only close in a pointwise sense throughout the closed subset . We require only that the subset be closed; in particular, the case that is finite is covered by our results. The restriction that is sharp in the sense that our result is false for .
期刊介绍:
The Journal of the London Mathematical Society has been publishing leading research in a broad range of mathematical subject areas since 1926. The Journal welcomes papers on subjects of general interest that represent a significant advance in mathematical knowledge, as well as submissions that are deemed to stimulate new interest and research activity.