<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Variance-Reduction on Nikita Podlozhniy</title>
    <link>https://npodlozhniy.github.io/tags/variance-reduction/</link>
    <description>Recent content in Variance-Reduction on Nikita Podlozhniy</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-GB</language>
    <lastBuildDate>Tue, 01 Oct 2024 00:00:00 +0000</lastBuildDate><atom:link href="https://npodlozhniy.github.io/tags/variance-reduction/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Linearization: Turning Ratio Metrics into Per-User Metrics</title>
      <link>https://npodlozhniy.github.io/posts/linearization/</link>
      <pubDate>Tue, 01 Oct 2024 00:00:00 +0000</pubDate>
      <guid>https://npodlozhniy.github.io/posts/linearization/</guid>
      <description>A simple algebraic transformation that converts ratio metrics into independent per-user values, making T-tests and sensitivity techniques directly applicable</description>
    </item>
    
    <item>
      <title>Delta Method for Ratio Metrics in AB Testing</title>
      <link>https://npodlozhniy.github.io/posts/delta-method/</link>
      <pubDate>Wed, 31 Jul 2024 00:00:00 +0000</pubDate>
      <guid>https://npodlozhniy.github.io/posts/delta-method/</guid>
      <description>Correct variance estimation for ratio metrics using the Delta Method — and why naive approaches quietly break your confidence intervals</description>
    </item>
    
    <item>
      <title>CUPED: Reducing Metric Variance with Pre-Experiment Data</title>
      <link>https://npodlozhniy.github.io/posts/cuped/</link>
      <pubDate>Thu, 18 Apr 2024 00:00:00 +0000</pubDate>
      <guid>https://npodlozhniy.github.io/posts/cuped/</guid>
      <description>How CUPED uses pre-experiment covariates to cut metric variance and boost AB test sensitivity — with a rigorous proof of what works and what silently breaks</description>
    </item>
    
    <item>
      <title>Bootstrap Methods for AB Testing</title>
      <link>https://npodlozhniy.github.io/posts/bootstrap/</link>
      <pubDate>Tue, 16 Aug 2022 00:00:00 +0000</pubDate>
      <guid>https://npodlozhniy.github.io/posts/bootstrap/</guid>
      <description>Classical and Poisson bootstrap as non-parametric tools for normalizing skewed metrics and enabling valid T-tests on small samples</description>
    </item>
    
    <item>
      <title>Bucketing: Variance Reduction and Faster AB Tests</title>
      <link>https://npodlozhniy.github.io/posts/bucketing/</link>
      <pubDate>Tue, 16 Aug 2022 00:00:00 +0000</pubDate>
      <guid>https://npodlozhniy.github.io/posts/bucketing/</guid>
      <description>How aggregating observations into buckets normalises skewed metrics, reduces variance, and cuts computation — with a probability proof for choosing the right bucket count</description>
    </item>
    
  </channel>
</rss>
