<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>Detrend — detrending and trend significance</title>
    <link>https://detrend.datacorp.top/</link>
    <description>Browser-based detrending, robust denoising and Mann-Kendall trend analysis with independent statistical verification.</description>
    <language>en</language>
    <copyright>© 2026 Detrend</copyright>
    <managingEditor>jizhi2024@foxmail.com (Detrend)</managingEditor>
    <lastBuildDate>Thu, 19 Nov 2026 00:00:00 GMT</lastBuildDate>
    <ttl>1440</ttl>
    <atom:link href="https://detrend.datacorp.top/rss.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>Detrend — browser-based detrending tool</title>
      <link>https://detrend.datacorp.top/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/</guid>
      <description>A detrending tool for Excel: diagnose stationarity, denoise robustly, run Mann-Kendall trend tests and every number against an independent oracle. No upload.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Product</category>
    </item>
    <item>
      <title>Detrending methods: STL, MK and robust denoising</title>
      <link>https://detrend.datacorp.top/methods/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/methods/</guid>
      <description>Method notes for the workspace: stationarity diagnostics, STL and differencing, Mann-Kendall corrections, and IQR, MAD and Qn robust denoising in the browser.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>Detrending methods: STL, differencing, residuals</title>
      <link>https://detrend.datacorp.top/methods/detrending/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/methods/detrending/</guid>
      <description>Compare STL decomposition, first-order and seasonal differencing and regression residual detrending, with unit-root diagnostics and the failure modes.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>Mann-Kendall trend test: methods and corrections</title>
      <link>https://detrend.datacorp.top/methods/mann-kendall/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/methods/mann-kendall/</guid>
      <description>Standard MK, Yue–Wang correction, TFPW, SMK, MKRD and MKC3: applicability and failure modes, shown side by side with effect sizes and Theil–Sen intervals.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>STL decomposition: trend, season and residual</title>
      <link>https://detrend.datacorp.top/methods/stl-decomposition/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/methods/stl-decomposition/</guid>
      <description>STL parameter semantics, robust mode, large-sample degradation and how residual denoising and ACF diagnostics connect in a browser-based analysis workflow.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>Robust denoising: IQR, MAD or Qn</title>
      <link>https://detrend.datacorp.top/methods/robust-denoising/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/methods/robust-denoising/</guid>
      <description>IQR fences, modified Z-score, MAD versus Qn efficiency and breakdown point, when STL residual denoising fits time-series data, and their failure modes.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>Statistical verification: independent oracles</title>
      <link>https://detrend.datacorp.top/verification/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/verification/</guid>
      <description>Public verification matrix: every statistic maps to independent reference implementation, a preset tolerance and a deterministic fixture. Re-run the comparison.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Verification</category>
    </item>
    <item>
      <title>FAQ: detrending and trend testing</title>
      <link>https://detrend.datacorp.top/faq/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/faq/</guid>
      <description>Common questions about detrending safety, Mann-Kendall method selection, multi-sheet Excel compatibility, trend statistics and data privacy.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>FAQ</category>
    </item>
    <item>
      <title>Privacy: local data analysis, no upload</title>
      <link>https://detrend.datacorp.top/privacy/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/privacy/</guid>
      <description>File bytes never leave your device: no upload, no server-side analysis, no data at rest. Read what is stored, what is not, and how advertising is bounded.</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>en</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Privacy</category>
    </item>
    <item>
      <title>Detrend — 浏览器内去趋势与趋势检验</title>
      <link>https://detrend.datacorp.top/zh-CN/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/</guid>
      <description>面向 Excel 的去趋势工具：先做平稳性诊断，再稳健去噪，运行 Mann-Kendall 趋势检验，并将每个数字对照独立 oracle。无需上传。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Product</category>
    </item>
    <item>
      <title>去趋势方法：Mann-Kendall、STL 与稳健去噪</title>
      <link>https://detrend.datacorp.top/zh-CN/methods/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/methods/</guid>
      <description>Detrend 工作台的方法说明：平稳性诊断、STL 与差分、带自相关校正的 Mann-Kendall 变体，以及 IQR、MAD、Qn 去噪。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>去趋势方法：STL、差分与回归残差</title>
      <link>https://detrend.datacorp.top/zh-CN/methods/detrending/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/methods/detrending/</guid>
      <description>比较 STL 分解、一阶与季节差分、回归残差去趋势，并说明单位根诊断与关键失效模式。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>Mann-Kendall 趋势检验：方法与自相关校正</title>
      <link>https://detrend.datacorp.top/zh-CN/methods/mann-kendall/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/methods/mann-kendall/</guid>
      <description>标准 MK、Yue–Wang 校正、TFPW、SMK、MKRD 与 MKC3：适用边界与失效模式，并排展示效应量与 Theil-Sen 区间。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>STL 分解：趋势、季节与残差</title>
      <link>https://detrend.datacorp.top/zh-CN/methods/stl-decomposition/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/methods/stl-decomposition/</guid>
      <description>STL 参数语义、稳健模式、大样本降级，以及残差去噪与 ACF 诊断在浏览器分析流程中的连接方式。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>稳健去噪：IQR、MAD 还是 Qn</title>
      <link>https://detrend.datacorp.top/zh-CN/methods/robust-denoising/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/methods/robust-denoising/</guid>
      <description>IQR 围栏、修正 Z 分数、MAD 与 Qn 的效率与崩溃点，以及何时应使用 STL 残差去噪而不是横截面规则。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Methods</category>
    </item>
    <item>
      <title>统计验证矩阵：独立 oracle</title>
      <link>https://detrend.datacorp.top/zh-CN/verification/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/verification/</guid>
      <description>公开验证矩阵：每个统计量映射到独立参考实现、预设容差与确定性 fixture。任何人都可以重跑对照。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Verification</category>
    </item>
    <item>
      <title>常见问题：去趋势与趋势检验</title>
      <link>https://detrend.datacorp.top/zh-CN/faq/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/faq/</guid>
      <description>关于去趋势安全性、Mann-Kendall 方法选择、多工作表 Excel 兼容性、趋势统计与数据隐私的常见问题。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>FAQ</category>
    </item>
    <item>
      <title>隐私：本地数据分析，不上传</title>
      <link>https://detrend.datacorp.top/zh-CN/privacy/</link>
      <guid isPermaLink="true">https://detrend.datacorp.top/zh-CN/privacy/</guid>
      <description>文件字节不离开设备：无上传、无服务端分析、无静态数据。本页说明存储了什么、没有存储什么，以及第三方广告处在什么位置。</description>
      <pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate>
      <dc:language>zh-CN</dc:language>
      <dc:creator>Detrend</dc:creator>
      <category>Privacy</category>
    </item>
  </channel>
</rss>