<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic QQ Plot behavior under factor coding and no-intercept model in Discussions</title>
    <link>https://community.jmp.com/t5/Discussions/QQ-Plot-behavior-under-factor-coding-and-no-intercept-model/m-p/910506#M106954</link>
    <description>&lt;P data-start="178" data-end="192"&gt;Hi everyone,&lt;/P&gt;
&lt;P data-start="197" data-end="413"&gt;When I fit a model in JMP using &lt;STRONG data-start="229" data-end="246"&gt;factor coding&lt;/STRONG&gt; and set the model to &lt;STRONG data-start="268" data-end="284"&gt;no intercept&lt;/STRONG&gt;, I noticed that the &lt;STRONG data-start="305" data-end="348"&gt;residual normal quantile plot (QQ plot)&lt;/STRONG&gt; behaves differently compared to the version with an intercept.&lt;/P&gt;
&lt;P data-start="418" data-end="821"&gt;Specifically, the &lt;STRONG data-start="436" data-end="482"&gt;residuals and the fitted “robust fit line”&lt;/STRONG&gt; in the QQ plot seem to &lt;STRONG data-start="506" data-end="526"&gt;shift vertically&lt;/STRONG&gt;, and even after centering the residuals, the &lt;STRONG data-start="572" data-end="617"&gt;slope of the fitted line changes slightly&lt;/STRONG&gt;. In addition, I also noticed that in some cases, the &lt;STRONG data-start="671" data-end="716"&gt;larger residuals (right side of the plot)&lt;/STRONG&gt; still lie close to the fitted line, while the &lt;STRONG data-start="763" data-end="796"&gt;smaller residuals (left side)&lt;/STRONG&gt; deviate more strongly.&lt;/P&gt;
&lt;P data-start="826" data-end="993"&gt;I’m wondering how JMP internally handles residuals and the “robust fit line” calculation when the model is fit &lt;STRONG data-start="937" data-end="961"&gt;without an intercept&lt;/STRONG&gt; and factors are effect coded.&lt;/P&gt;
&lt;P data-start="998" data-end="1143"&gt;Could anyone explain why this shift and slope difference occur, and how the QQ plot line and confidence curves are computed under this setting?&lt;/P&gt;
&lt;P data-start="1148" data-end="1177"&gt;Thanks a lot for your help!&lt;/P&gt;
&lt;P data-start="1148" data-end="1177"&gt;JMP PRO 17.1.0&lt;/P&gt;</description>
    <pubDate>Thu, 30 Oct 2025 11:56:16 GMT</pubDate>
    <dc:creator>happy</dc:creator>
    <dc:date>2025-10-30T11:56:16Z</dc:date>
    <item>
      <title>QQ Plot behavior under factor coding and no-intercept model</title>
      <link>https://community.jmp.com/t5/Discussions/QQ-Plot-behavior-under-factor-coding-and-no-intercept-model/m-p/910506#M106954</link>
      <description>&lt;P data-start="178" data-end="192"&gt;Hi everyone,&lt;/P&gt;
&lt;P data-start="197" data-end="413"&gt;When I fit a model in JMP using &lt;STRONG data-start="229" data-end="246"&gt;factor coding&lt;/STRONG&gt; and set the model to &lt;STRONG data-start="268" data-end="284"&gt;no intercept&lt;/STRONG&gt;, I noticed that the &lt;STRONG data-start="305" data-end="348"&gt;residual normal quantile plot (QQ plot)&lt;/STRONG&gt; behaves differently compared to the version with an intercept.&lt;/P&gt;
&lt;P data-start="418" data-end="821"&gt;Specifically, the &lt;STRONG data-start="436" data-end="482"&gt;residuals and the fitted “robust fit line”&lt;/STRONG&gt; in the QQ plot seem to &lt;STRONG data-start="506" data-end="526"&gt;shift vertically&lt;/STRONG&gt;, and even after centering the residuals, the &lt;STRONG data-start="572" data-end="617"&gt;slope of the fitted line changes slightly&lt;/STRONG&gt;. In addition, I also noticed that in some cases, the &lt;STRONG data-start="671" data-end="716"&gt;larger residuals (right side of the plot)&lt;/STRONG&gt; still lie close to the fitted line, while the &lt;STRONG data-start="763" data-end="796"&gt;smaller residuals (left side)&lt;/STRONG&gt; deviate more strongly.&lt;/P&gt;
&lt;P data-start="826" data-end="993"&gt;I’m wondering how JMP internally handles residuals and the “robust fit line” calculation when the model is fit &lt;STRONG data-start="937" data-end="961"&gt;without an intercept&lt;/STRONG&gt; and factors are effect coded.&lt;/P&gt;
&lt;P data-start="998" data-end="1143"&gt;Could anyone explain why this shift and slope difference occur, and how the QQ plot line and confidence curves are computed under this setting?&lt;/P&gt;
&lt;P data-start="1148" data-end="1177"&gt;Thanks a lot for your help!&lt;/P&gt;
&lt;P data-start="1148" data-end="1177"&gt;JMP PRO 17.1.0&lt;/P&gt;</description>
      <pubDate>Thu, 30 Oct 2025 11:56:16 GMT</pubDate>
      <guid>https://community.jmp.com/t5/Discussions/QQ-Plot-behavior-under-factor-coding-and-no-intercept-model/m-p/910506#M106954</guid>
      <dc:creator>happy</dc:creator>
      <dc:date>2025-10-30T11:56:16Z</dc:date>
    </item>
  </channel>
</rss>

