<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Probability and Statistics - Medium]]></title>
        <description><![CDATA[Interesting examples on Probability and Statistics - Medium]]></description>
        <link>https://medium.com/probability-and-statistics?source=rss----f29e7d1a6572---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Probability and Statistics - Medium</title>
            <link>https://medium.com/probability-and-statistics?source=rss----f29e7d1a6572---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 11:31:04 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/probability-and-statistics" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Monty Hall Problem — Bayesian Reasoning]]></title>
            <link>https://medium.com/probability-and-statistics/monty-hall-problem-bayesian-reasoning-d9caf44c194b?source=rss----f29e7d1a6572---4</link>
            <guid isPermaLink="false">https://medium.com/p/d9caf44c194b</guid>
            <category><![CDATA[bayesian]]></category>
            <category><![CDATA[puzzle]]></category>
            <category><![CDATA[probability]]></category>
            <dc:creator><![CDATA[yogesh luthra]]></dc:creator>
            <pubDate>Sun, 15 Oct 2017 12:00:17 GMT</pubDate>
            <atom:updated>2017-10-15T11:51:52.033Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1008/1*WdSd9bt_VtYXTOxsySX5kA.jpeg" /></figure><p>Although it is a very popular problem, it still puzzles many people. I find Bayesian reasoning gives a simple and intuitive solution.</p><p>Problem description: “Suppose you’re on a game show, and you’re given the choice of three doors: Behind one door is a car; behind the others, goats. You pick a door, say №1, and the host, who knows what’s behind the doors, opens another door, say №3, which has a goat. He then says to you, “Do you want to pick door №2?” Is it to your advantage to switch your choice?</p><p>Visualize three doors:</p><pre>&quot;1&quot;     &quot;2&quot;     &quot;3&quot;</pre><p>We are basically interested in finding <strong>p(winning car | strategy)</strong> (that is, probability of winning car, given a strategy).</p><p>Now we simply have 2 strategies.</p><ol><li>Choose a door at random and then <strong>stick</strong> to it when host asks.</li><li>Choose a door at random and <strong>switch</strong> when host asks.</li></ol><p>With no knowledge of where the car is, <strong>prior probability</strong> of car behind any door is 1/3.</p><h3>Strategy: Switch</h3><pre>p(winning car | Switch) = <br>p(winning car, car behind &quot;1&quot; | Switch) + <br>p(winning car, car behind &quot;2&quot; | Switch) + <br>p(winning car, car behind &quot;3&quot; | Switch) ..... (marginalization)</pre><pre>=<br>p(winning car | Switch, car behind &quot;1&quot;)*p(car behind &quot;1&quot; | Switch) + <br>p(winning car | Switch, car behind &quot;2&quot;)*p(car behind &quot;2&quot; | Switch) + <br>p(winning car | Switch, car behind &quot;3&quot;)*p(car behind &quot;3&quot; | Switch) ... (chain rule)<br>Note: p(car behind a door) is independent of your strategy. <br>i.e. p(car behind &quot;1&quot; | Stick) = p(car behind &quot;1&quot;)</pre><p>Say you choose door “1” initially, but will switch when host asks.</p><p>p(winning car | Switch, car behind “1”) = 0</p><p><strong>p(winning car | Switch, car behind “2”) = 1 … why? </strong>because you initially chose “1” and if car is behind “2”, host will obviously open “3”. If you switch (of-course to “2” now) then Bang! you have the car.</p><p>p(winning car | Switch, car behind “3”) = 1 … for same reason as above.</p><p>So,</p><p><strong>p(winning car | Switch) = 0*1/3 + 1*1/3 + 1*1/3 = 2/3</strong></p><h3>Strategy: Stick</h3><pre>p(winning car | Stick) = <br>p(winning car, car behind &quot;1&quot; | Stick) + <br>p(winning car, car behind &quot;2&quot; | Stick) + <br>p(winning car, car behind &quot;3&quot; | Stick)</pre><pre>= <br>p(winning car | Stick, car behind &quot;1&quot;)*p(car behind &quot;1&quot; | Stick) + <br>p(winning car | Stick, car behind &quot;2&quot;)*p(car behind &quot;2&quot; | Stick) + <br>p(winning car | Stick, car behind &quot;3&quot;)*p(car behind &quot;3&quot; | Stick)</pre><p>You will make only one choice of door, say “1” initially and if you stick to it, then:</p><p><strong>p(winning car | Stick) = 1 * 1/3 + 0*1/3 + 0*1/3 = 1/3</strong></p><h3>Summary:</h3><p>Why is winning probability higher when Strategy is “Switch”? Basically 2 things:</p><ol><li>You are likely to choose right door with only 1/3 probability.</li><li>That is, you choose wrong door with 2/3 probability and then host gives you a major hint of which door you should actually switch to!</li></ol><p>BTW, this was a real show and real cars were given away for many years.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d9caf44c194b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/probability-and-statistics/monty-hall-problem-bayesian-reasoning-d9caf44c194b">Monty Hall Problem — Bayesian Reasoning</a> was originally published in <a href="https://medium.com/probability-and-statistics">Probability and Statistics</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>