<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>ARIMA on Quick Insights</title>
    <link>https://www.yakauleu.com/tags/arima/</link>
    <description>Recent content in ARIMA on Quick Insights</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-us</language>
    <copyright>© 2026 Yauheni Yakauleu</copyright>
    <lastBuildDate>Fri, 20 Nov 2026 09:00:00 +0000</lastBuildDate><atom:link href="https://www.yakauleu.com/tags/arima/index.xml" rel="self" type="application/rss+xml" />
    
    <item>
      <title>AR(1) Process Autocorrelation: Theory vs Simulation in Python</title>
      <link>https://www.yakauleu.com/post/ar1_process_autocorrelation_verification/</link>
      <pubDate>Fri, 20 Nov 2026 09:00:00 +0000</pubDate>
      
      <guid>https://www.yakauleu.com/post/ar1_process_autocorrelation_verification/</guid>
      <description>&lt;p&gt;You simulate an AR(1) process with parameter &lt;code&gt;ρ = 0.7&lt;/code&gt; and compute its sample autocorrelation at lag 1. What value do you expect? The theoretical answer is simple: the lag-1 autocorrelation of an AR(1) process is exactly &lt;code&gt;ρ&lt;/code&gt;. But how quickly does the sample autocorrelation converge to this value — and how noisy is the estimator at realistic sample sizes?&lt;/p&gt;&#xA;&lt;p&gt;In this post, we derive the theoretical autocorrelation function of an AR(1) process, simulate it in NumPy, and verify both the &lt;em&gt;point estimate&lt;/em&gt; and its &lt;em&gt;dispersion&lt;/em&gt; against theory with a Monte-Carlo experiment.&lt;/p&gt;</description>
      
    </item>
    
  </channel>
</rss>
