<?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[Stories by Harry Izatt on Medium]]></title>
        <description><![CDATA[Stories by Harry Izatt on Medium]]></description>
        <link>https://medium.com/@h.j.izatt?source=rss-a898ece038a0------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*6xfiPwuGqaHEnCtG.jpg</url>
            <title>Stories by Harry Izatt on Medium</title>
            <link>https://medium.com/@h.j.izatt?source=rss-a898ece038a0------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 13:11:13 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@h.j.izatt/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[When’s the best time for a vacation in Seattle?]]></title>
            <link>https://medium.com/@h.j.izatt/whens-the-best-time-for-a-vacation-in-seattle-e035f17800e2?source=rss-a898ece038a0------2</link>
            <guid isPermaLink="false">https://medium.com/p/e035f17800e2</guid>
            <category><![CDATA[travel]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[airbnb]]></category>
            <category><![CDATA[seattle]]></category>
            <dc:creator><![CDATA[Harry Izatt]]></dc:creator>
            <pubDate>Thu, 20 Oct 2022 22:25:29 GMT</pubDate>
            <atom:updated>2022-10-20T22:33:24.187Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Y3TBJGSP7qLUSsoZ.jpg" /></figure><p>Everyone loves a good vacation, and Airbnb is a great place to hunt for a bargain. Famed for its <a href="https://twitter.com/dadsof3/status/1581989078408790020?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1581989078408790020%7Ctwgr%5E846f3a001cf490154282d4c944b5b15753d9d298%7Ctwcon%5Es1_&amp;ref_url=https%3A%2F%2Fwww.buzzfeed.com%2Falexalisitza%2Fairbnb-bookings-down-twitter-reacts">low prices</a>, <a href="https://www.buzzfeed.com/kristatorres/hidden-camera-airbnb-tiktok">respect for privacy</a>, and <a href="https://dailyhive.com/vancouver/airbnb-chores-list">deep culture of hospitality</a>, the company has also published a great deal of data on its listings in multiple cities. Just in August I took a trip to a Seattle Airbnb property, and so I thought I’d take a dive into their <a href="https://www.kaggle.com/datasets/airbnb/seattle">Seattle open data</a>.</p><p>The data spans from January 2016 to January 2017. I’m most interested in calendar.csv, which covers the anonymized identities of listings, given dates, availabilities, and when available, the nightly cost. I’m interested in two questions:</p><ol><li><strong>Which months have the greatest availability?</strong></li><li><strong>How do prices rise or fall throughout the year?</strong></li></ol><h4>Question 1: Which months have the greatest availability?</h4><p>I took a few humorous missteps with data visualization — the dataset is so large and dense (1,393,570 entries!) that conventional scatter plots yield an uninterpretable blob of dots. Limiting to entries priced below $500/night does little to help.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/961/1*OTqeRnJuHi9LU27uHVz7lA.png" /><figcaption>How insightful</figcaption></figure><p>I had more luck totalling across months.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lydaxFX1n882WscvLn28rg.png" /></figure><p>Fig 1 shows all listings in the dataset, available or not, while Fig 2 is just those available with a price listed. It doesn’t look like there’s a surge in units posted on the market in the summer vacation season — in fact, June through August have fewer available listings than the months before and after. Figure 2 is much more variable, suggesting hosts may keep their units on Airbnb’s website but not actively posted for guests.</p><p>The miniscule availability for January 2017 stood out to me. Checking the data, the latest date included in the dataset is January 2nd, so this month really just having one night included explains why its value is so low. It does raise the question of why they’d bother including a single night from 2017 in a dataset otherwise entirely concerned with 2016.</p><h4>Question 2: How do prices vary over time?</h4><p>I also ran into some troubles with visualizing this data in a basic scatterplot due to the many data points.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/680/1*BDPrzT3I4EATQjskg4Dohg.png" /></figure><p>So I opted for a histogram. This splits the daset into 75 price brackets, and plots the number of entries in each. The higher the bar, the more listings in the price range.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qcmddl7GrI20ykP8sDffIA.png" /></figure><p>Fig 1 shows the whole range, Fig 2 shows just those under $500/night, and Fig 3 shows those above $500. As we see, it has a rough bell-shape distribution, but no prices below $0, and a long “right tail” at the upper price range. Fig 2 zooms in on just those in a more normal price range. We can see most listings are around $100/night, with few beyond $200. Prices range from as little as $10 all the way to $1650.</p><p>It’s important to note that the y-axis values are different for each plot: Fig 1 peaks at 160,000, Fig 2 around 55,000 and Fig 3 just above 2000; there isn’t a huge spike in listings around $600!</p><p>Prices also vary over time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/375/1*4nLsWSUZj1YB3eqCi18nBw.png" /></figure><p>While there were fewer availabilities in the summer season, this time sees the highest overall prices. Prices fall through the autumn months, but still end above the starting level in January.</p><p>To do a more rigorous analysis of the price change, I ran a series of linear regressions. Put simply, these find the “best fit” between an outcome variable (price) and one or more explanatory variables (time through the year).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R4COHe494SYYBHCTSeAdcA.png" /></figure><p>Fig 5 is just linear — it shows how prices rise on average with the passage of each month. Fig 6 fits price to both linear months elapsed, and months elapsed squared, leading to the nonlinear curve. Visually, Fig 5 fits the data much better, accounting for prices rising and falling over the year. “Adjusted R2” describes how much of the data’s variation a model accurately explains. The models’ values suggest Fig 5 explains just 20.5% of the data’s variation, while Fig 6 explains 78.5% — a fairly good fit.</p><p>However, we need to use caution when interpreting the results. Linear regressions cannot be simply extended to new data outside of the sample. In other words, our models can be used to predict the average values of Seattle Airbnb listings from January to December of 2016, but cannot reliably extend these predictions back into 2015 or forward into 2017.</p><h4>Conclusion</h4><p>On my <a href="https://github.com/HJIzt/DataScienceBlogpost/blob/master/seattle_airbnb_analysis.ipynb">Github</a>, I have more failed and successful plots, as well as direct model results, and an investigation into whether the price variation is from hosts raising prices of units already on the market, or if it’s from high-price units being made available in the summer.</p><p>(Long story short, it’s both.)</p><p>The main takeaway is that at least as far as lodging availability and prices go, the best time to visit Seattle is in December and January. Which I’d definitely recommend. The Pacific Northwest is lovely that time of year.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*Qg3nWM8WJjrLaNzm.jpg" /><figcaption>Bring your raincoat and Blundstones.</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e035f17800e2" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>