<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0"
    xmlns:content="http://purl.org/rss/1.0/modules/content/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
        <title>GLM.ModelRefinement — BattleActs Exam 8 Forum</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/</link>
        <pubDate>Sun, 05 Apr 2026 18:06:51 +0000</pubDate>
        <language>en</language>
            <description>GLM.ModelRefinement — BattleActs Exam 8 Forum</description>
    <atom:link href="https://www.battleacts8.ca/8/forum/index.php?p=/categories/glm-modelrefinement/feed.rss" rel="self" type="application/rss+xml"/>
    <item>
        <title>Gini Index</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/discussion/275/gini-index</link>
        <pubDate>Fri, 22 Aug 2025 15:19:11 +0000</pubDate>
        <category>GLM.ModelRefinement</category>
        <dc:creator>Lcparga1</dc:creator>
        <guid isPermaLink="false">275@/8/forum/index.php?p=/discussions</guid>
        <description><![CDATA[<p>Hello,</p><p>Does having a higher Gini Index mean the rating plan is able to differentiate the best and worst risks better than a rating plan that has a smaller Gini index?</p><p>Thanks in advance!</p>]]>
        </description>
    </item>
    <item>
        <title>Excel Problem: GLM_ModelRefinement1</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/discussion/253/excel-problem-glm-modelrefinement1</link>
        <pubDate>Tue, 27 Aug 2024 12:23:05 +0000</pubDate>
        <category>GLM.ModelRefinement</category>
        <dc:creator>kmo484</dc:creator>
        <guid isPermaLink="false">253@/8/forum/index.php?p=/discussions</guid>
        <description><![CDATA[<p>I may be missing something obvious, but can you explain where the $45,000 comes from in the solution for part a?  How did we know it was the upper range of Vehicle Value?</p>]]>
        </description>
    </item>
    <item>
        <title>Unscaled Deviance is independent</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/discussion/139/unscaled-deviance-is-independent</link>
        <pubDate>Tue, 18 Oct 2022 23:01:54 +0000</pubDate>
        <category>GLM.ModelRefinement</category>
        <dc:creator>jmoore13</dc:creator>
        <guid isPermaLink="false">139@/8/forum/index.php?p=/discussions</guid>
        <description><![CDATA[<p>I'm confused by why exactly Goldburd uses the term "independent" when talking about the dispersion parameter in unscaled deviance.  Unscaled Deviance literally is reliant on the value of the dispersion parameter in the formula, so by definition it isn't independent and depends on what the dispersion parameter is. </p><p>I get that they're saying without the dispersion parameter, we couldn't compare distributions since Scaled Deviance doesn't account for difference in distributions andthe dispersion paraemter brings that in, but is there a reason they're using "independent" that I'm not understanding?</p>]]>
        </description>
    </item>
    <item>
        <title>Binned residuals</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/discussion/88/binned-residuals</link>
        <pubDate>Thu, 18 Aug 2022 14:22:00 +0000</pubDate>
        <category>GLM.ModelRefinement</category>
        <dc:creator>emelnd</dc:creator>
        <guid isPermaLink="false">88@/8/forum/index.php?p=/discussions</guid>
        <description><![CDATA[<p>When you bin the residuals into groups with the same sum of working weights, is the process order by working residuals and keep adding until you get close to the target size, or do you combine random data points to get to the target size regardless of the working residual value?</p>]]>
        </description>
    </item>
    <item>
        <title>Hinge function</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/discussion/87/hinge-function</link>
        <pubDate>Wed, 17 Aug 2022 15:30:11 +0000</pubDate>
        <category>GLM.ModelRefinement</category>
        <dc:creator>emelnd</dc:creator>
        <guid isPermaLink="false">87@/8/forum/index.php?p=/discussions</guid>
        <description><![CDATA[<p>What's a hinge function?</p>]]>
        </description>
    </item>
    <item>
        <title>2018 Fall Q6 F Statistics</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/discussion/31/2018-fall-q6-f-statistics</link>
        <pubDate>Wed, 11 Aug 2021 20:10:34 +0000</pubDate>
        <category>GLM.ModelRefinement</category>
        <dc:creator>jiangz12</dc:creator>
        <guid isPermaLink="false">31@/8/forum/index.php?p=/discussions</guid>
        <description><![CDATA[<p>Hello, </p><p>Based on the F Statistic formula, The numerator is the <strong>Unscaled Deviance</strong> =phi * 2*(ll sat - ll model) . </p><div>
    <div>
        <a href="https://battleacts8.ca/8/forum/uploads/821/U7IM7X2434EU.png" rel="nofollow noreferrer noopener ugc" target="_blank">
            <img src="https://battleacts8.ca/8/forum/uploads/821/U7IM7X2434EU.png" alt="image.png" height="82" width="454" /></a>
    </div>
</div>
<p>Therefore in this question Ds = 2(-1000-(-1500))*1.75 = 1750, and Db = 2(-1000-(-1465))*1.75 = 1627.5</p><p>and F = (1750-1627.5)/(2*1.75) = 35</p><p>The sample answer seem to be using the <strong>scaled deviance</strong>, leading to an answer of 20. Would you please advise which is correct, the paper or the question? Or is there a reason why the scaled deviance is used in this question instead of the unscaled deviance. </p><p><br /></p><div>
    <div>
        <a href="https://battleacts8.ca/8/forum/uploads/622/C6RGJG7P0RB1.png" rel="nofollow noreferrer noopener ugc" target="_blank">
            <img src="https://battleacts8.ca/8/forum/uploads/622/C6RGJG7P0RB1.png" alt="image.png" height="258" width="777" /></a>
    </div>
</div>
<p>Thank you lots!</p>]]>
        </description>
    </item>
    <item>
        <title>F17 Q5 - Quintile Plots</title>
        <link>https://www.battleacts8.ca/8/forum/index.php?p=/discussion/7/f17-q5-quintile-plots</link>
        <pubDate>Sat, 19 Jun 2021 16:03:59 +0000</pubDate>
        <category>GLM.ModelRefinement</category>
        <dc:creator>stratthixx</dc:creator>
        <guid isPermaLink="false">7@/8/forum/index.php?p=/discussions</guid>
        <description><![CDATA[<p>I think this question is in the wrong battle quiz. The material is covered in a later section and should be in Battle Quiz #2. If I am wrong, could you point me to where I could find the material? Thanks.</p>]]>
        </description>
    </item>
   </channel>
</rss>
