<?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 Emi Harry on Medium]]></title>
        <description><![CDATA[Stories by Emi Harry on Medium]]></description>
        <link>https://medium.com/@Emi.N.Harry?source=rss-df2627a77831------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*ZwwxTk8xth4e5gMdASQh6w.jpeg</url>
            <title>Stories by Emi Harry on Medium</title>
            <link>https://medium.com/@Emi.N.Harry?source=rss-df2627a77831------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 18:53:48 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@Emi.N.Harry/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[Is the American Dream Attainable by All?]]></title>
            <link>https://medium.com/@Emi.N.Harry/is-the-american-dream-attainable-by-all-9d7a7b275f66?source=rss-df2627a77831------2</link>
            <guid isPermaLink="false">https://medium.com/p/9d7a7b275f66</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[blacklivesmatter]]></category>
            <category><![CDATA[home-loan]]></category>
            <dc:creator><![CDATA[Emi Harry]]></dc:creator>
            <pubDate>Sat, 20 Jun 2020 03:41:31 GMT</pubDate>
            <atom:updated>2020-06-20T04:07:10.793Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/510/1*C-I8OS86h3nlIJhMWHfxNg.jpeg" /></figure><p>Today being JUNETEENTH, it feels appropriate to start a series exploring <a href="https://csce.ucmss.com/cr/books/2018/LFS/CSREA2018/ICD8073.pdf">past</a> and present research findings, on the state of financial lending practices towards minorities.</p><p>Brief history: On September 22, 1862, President <a href="https://en.wikipedia.org/wiki/Abraham_Lincoln">Abraham Lincoln</a> made the proclamation granting all African American slaves their freedom, which became effective on January 1, 1863, but it wasn’t until June 19th 1865, that African Americans in Texas gained their freedom, hence <strong>JUNETEENTH</strong>.</p><p>Nevertheless, life wasn’t rosy for these newly freed slaves as they had to earn wages to survive but were often left without. An example of such was the enactment of redlining. The term “Redlining” has been used over the years to describe a perpetual form of discrimination; where financial institutions refused to provide their services to neighborhoods they termed as “African-American” neighborhoods.</p><p>According to a publication by Brent Gaspaire in BlackPast.org, the origin of the term stems from the policies developed by the Home Owners Loan Corporation (HOLC) created in 1933, by the Franklin Roosevelt Administration, to reduce home foreclosures during the Depression, and then institutionalized by the 1937 U.S. Housing Act which established the Federal Housing Association (FHA).</p><p>Federal housing agencies including the HOLC, and the FHA determined whether areas were deemed unfit for investment by banks, insurance companies, savings and loan associations, and other financial services companies. The areas were physically demarcated with red shading on a map. In contrast, zones which were to receive preferential lending status were marked in green shading and intermediate areas in blue shading. Often these decisions were arbitrarily based on the area’s racial composition rather than income levels. Despite the abolishment of “redlining” in the 1968 Civil Acts law that was passed, it continued. Neighborhoods that local/community banks deemed unfit for investment were left underdeveloped or in disrepair.</p><p>Now for those of you who are yet to watch ‘<a href="https://www.imdb.com/title/tt6285944/">The Banker</a>’, a movie about <strong>Bernard S. Garret and Joe Morris, </strong>the first African Americans to own banks, I recommend you watch it. Bernard’s activities in real estate, laid the foundation for the Home Mortgage Disclosure Act (HMDA), which was approved by congress in 1975.</p><p>The Act requires mortgage lenders to keep records of vital pieces of information regarding their lending practices (in other words, information on income, family size, the loan amount, purpose of the loan, the race of the applicant, marital status etc.), which they must submit to regulatory authorities. A crucial component in this Act, is Regulation C, mandating the monitoring of the geographic targets of mortgage lenders, provide an identification mechanism for any predatory lending practices, and to provide reporting statistics on the mortgage market to the government<em>.</em></p><p>In February 2018, I came across an <a href="https://www.revealnews.org/article/for-people-of-color-banks-are-shutting-the-door-to-homeownership/">article</a> detailing the issues being faced by minorities when trying to take out mortgages for their home purchase, giving rise to my assumption that:</p><blockquote>“Race still plays a huge role in the decision-making process of home mortgage approvals, or denials, and it has not improved since the last major report on it, in 2014.”</blockquote><p>Thus, I set out to test the validity of that assumption, by analyzing the 2007 to 2016 HMDA data, publicly available through the <a href="https://www.consumerfinance.gov/data-research/hmda/historic-data/">Consumer Financial Protection Bureau</a>.</p><p>The first step in the analysis was to answer few questions.</p><ul><li><strong>What wasthe overall ratio of loan approvals to denials from 2012 to 2016?</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/583/1*zbbu7ep44F3H3_DvkZ_8Mg.png" /><figcaption><em>Fig 1. The total number of home mortgage approvals and denials from 2012 to 2016.</em></figcaption></figure><p>From figure 1 above, it is clear that the ratio of approvals to denials is approximately a 5 to 1 ratio, implying that 5 in every 6 mortgage applications received between 2012 and 2016 were approved.</p><ul><li><strong>Has there been a consistent increase in the loan approvals from 2012 to 2016?</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/958/1*sScGHkcQOqwBjIEc8K0s7A.png" /><figcaption><em>Fig 2. The trend of home mortgage approvals and denials from 2012 to 2016.</em></figcaption></figure><p>Looking at the trend lines of both the number of denials and approvals by year, in figure 2 above, it wasn’t surprising to see that even three to five years after the 2009 recession, there was still a constant decline in the numbers, with it taking a huge dip in 2014. But by 2015 the numbers begin to increase. As of 2016 it was still not as high as it was in 2012. This means that overall, the number of individuals that sought mortgages continued to decrease even three to five years after the 2009 recession.</p><p>Given the premise of historical lending discrimination, the next question that rose was: how this has affected it? Leading to the final question.</p><ul><li><strong>What was the trend in approvals and denials by race from 2012 to 2016?</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fEfE7SThgtrDidoCnICLKw.png" /><figcaption><em>Fig 3. The trend of home mortgage approvals and denials from 2012 to 2016.</em></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JQwo0TAlcRdiLwF6SuLMBA.png" /><figcaption><em>Fig 4. The trend of home mortgage approvals and denials from 2012 to 2016.</em></figcaption></figure><p>From figure 3 and 4, it was deduced that in general there are always more white applicants than black applicants. <strong>Nevertheless, the data showed that approximately 85% of all white applicants get approved for a home loan, significantly higher than the 68% approval rate of the black applicants.</strong></p><p>In 2016 particularly, a total of 2,800,327 white applicants were accounted for in the data set, out of which, only 429,741 applications were denied, showing that white applicants had an approximate approval rate of 85%. Whereas, out of 156,185 black applicants, 51,231 applications were denied, making it an approximate approval rate of 67% in 2016.</p><p>Thus, while the ‘white approval rate” did not deviate from the overall average in 2016, the “black approval rate” <strong>declined by 1% </strong>from its average in that same year.</p><p>With the stats taken into consideration, I proceeded to train two machine learning models, the first was with a C5.0 decision tree model and the second was a neural network model.</p><p>The results were as follows:</p><blockquote>In the C5.0 model, <strong>approximately 32% </strong>of the black applicants who were approved for a home loan were miss-classified as white, while only <strong>18% </strong>of the white applicants who were approved for a home loan were misclassified as black. In the case of the denials, <strong>approximately 11%</strong> of the black applicants who were denied a home loan were misclassified as white, while <strong>approximately 40% </strong>of the white applicants who were denied a home loan were misclassified as black.</blockquote><blockquote>In the neural network model, <strong>approximately 41%</strong> of the black applicants who were approved for a home loan were misclassified as white, while only <strong>24% </strong>of the white applicants who were approved for a home loan were misclassified as black. In the case of the denials, <strong>approximately 12%</strong> of the black applicants who were denied a home loan were misclassified as white, while<strong> approximately 50%</strong> of the white applicants who were denied a home loan were misclassified as black.</blockquote><p>Given that in both models, the rate of misclassification of black applicants as white applicants, who had their loans approved, was approximately <strong>twice the rate</strong> of misclassification of white applicants as black applicants, who had their loans approved.</p><p>In addition, in both models, the rate of misclassification of white applicants as black applicants, who had their loans denied, was approximately <strong>four times the rate</strong> of misclassification of black applicants as white applicants, who had their loans denied.</p><p>It was highly conclusive that the assumption that: “<em> race still plays a huge role in the decision-making process of home mortgage approvals or denials, and it has not improved since the last major report on it in 2014” </em><strong>is true</strong>.</p><p>Based on that conclusion, it is safe to say that the ‘American dream’ doesn’t seem to be attainable by all.</p><p><strong>Note: </strong>New HMDA data relating to applicants is now being disclosed which contains important features like the applicant’s credit score, loan interest rate and spread, debt-to-income-ratio, and prepayment penalties. As such there’ll be a follow up to this story, taking a deeper dive to see how an applicant’s race determines things like the loan interest rate and spread.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9d7a7b275f66" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How To Earn Passive Income After COVID-19?]]></title>
            <link>https://medium.com/@Emi.N.Harry/how-to-earn-passive-income-after-covid-19-8fc9883f3559?source=rss-df2627a77831------2</link>
            <guid isPermaLink="false">https://medium.com/p/8fc9883f3559</guid>
            <category><![CDATA[new-york]]></category>
            <category><![CDATA[real-estate]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Emi Harry]]></dc:creator>
            <pubDate>Sat, 25 Apr 2020 21:27:40 GMT</pubDate>
            <atom:updated>2020-04-25T21:27:40.520Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1012/1*YTCCAE9DBhiofAYGF1dn7w.png" /></figure><p>As we all strive for financial independence, one question that seems to linger in the minds of most individuals is: how can I earn passive income?</p><p>I welcome you all to join me as I explore data available from different industries to see which are viable for the everyday individual, lacking a trust-fund to back such investments.</p><p>First up on the list, consider investing in real estate. Yes! you read right, I said real estate but don’t get me wrong, I’m not referring to flipping of houses or investing in massive buildings, which are all great if you have the money for it whether personal or loans. Instead, I’m referring to an investment in the shared real estate economy….Airbnb!</p><p>And no! I’m not an employee, neither am I a shareholder nor a paid promoter. I am a data scientist, and as they say the numbers never lie (while that maybe debatable in some cases, it holds true in most).</p><p>Thus, to determine if it is worth the hype, I set out to answers some questions through the <a href="https://www.kaggle.com/enharry/airbnb-compilation">Airbnb New York dataset</a>, and as some of you may know, New York City is currently their third biggest market.</p><p>First question…….</p><ul><li><strong><em>How has the number of bookings changed over the last 5 years?</em></strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/671/1*La6QhCdczdYWgO8hJpQloA.png" /><figcaption>Figure 1</figcaption></figure><p>While Airbnb was founded in 2008, they became operational in New York City, in 2009 and had legal battles in 2016.</p><p>Yet, figure 1, clearly shows that in the last 5 years, there has been a steady growth in the number of bookings through their platform. In 2015 they had approximately 311,360 bookings and by 2016 it grew by approximately 482%, to sit at 1,500,005 bookings.</p><p>Even though the growth rate declined consistently since then to 261%, 159%, and 115%, between 2017, 2018 and 2019 respectively, it is still remarkable.</p><p>From an investment standpoint, this shows resilience, and that’s a trait that is essential for any form of investment. In trying to understand why, the only explanation that holds true is human nature because in and out of global recessions, we seek personal gratification, and tourism is one way to achieve that.</p><p>This leads to the second question………</p><ul><li><strong><em>How does borough, and listing type affect pricing?</em></strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/812/1*5ueupE24VgnoCwA9Yvfl1w.png" /><figcaption>Figure 2</figcaption></figure><p>One does not need to be a rocket scientist to know that the nicer the neighborhood, the higher the price.</p><p>Nevertheless, if you are considering Airbnb as a source of passive income, knowing exactly how much your sofa-bed, or spare room is going for on the market is important.</p><p>Starting from the Bronx, the average market value for a private room is $70, which is surprisingly lower than the $95 average market value of a shared room. The same pattern is seen in Queens where the average market value for a private room is $76, while the average market value of a shared room is $92. Lastly, in Staten Island, where the average market value for a private room is $64, while the average market value of a shared room is $85.</p><p>Brooklyn and Manhattan on the other hand, follow the expected norm with the higher average market values of $77 and $110 respectively for a private room, and the lower average market values of $59 and $96 respectively for a shared room.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/832/1*yy-J3m7OYh9E5CihYqrz1g.png" /><figcaption>Figure 3</figcaption></figure><p>Though from looking at figure 3 above, the concept of subletting your entire apartment on Airbnb may seem very attractive due to the number of bookings, it is unreasonable. Here’s why; New York mandates that the only time you can list your apartment on the platform for anything less than 30 days is if you maintain it as your permanent residence.</p><p>As such, if an individual lives in an apartment and wants to sublet the entire apartment on Airbnb, the only way to sublet is if one is going away for 1 month. Thus, to take advantage of the <strong>‘Entire home/apt’</strong> listing investment opportunity, it is best to rent or buy a second apartment as an investment property.</p><p>Next up is seasonality…….</p><ul><li><strong><em>How does seasonality affect the total number of reservations?</em></strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R_3U3AwfDjwrlU5RPYb1Tg.png" /><figcaption>Figure 4</figcaption></figure><p>That’s a massive plot up there, but it’s important that you get the full picture (literally), and the summary of those subplots within is that over the last 5 years, <strong>Winter</strong> has been the lowest month for bookings in all 5 boroughs as is expected. By <strong>Spring</strong>, it beings to pick up, in all boroughs, and sustains into the <strong>Summer</strong>, with Bronx, Staten Island, and Queens showing a higher increase in bookings, at 109%, 108%, and 106% respectively, with Brooklyn and Manhattan trailing at 101% and 100% respectively. In the <strong>Fall</strong> however, while there’s a growth spike in most boroughs, the Bronx and Queens stall.</p><p>Now that we know in every season rooms are being rented, albeit at varying rates, let’s look at the lucrative neighborhoods across the city……..</p><ul><li><strong><em>Over the last 5 years, what are the top 10 listings based on the total number of bookings, price point and neighborhood?</em></strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EsbZ8oURrfVvdhsvOKlm4Q.png" /><figcaption>Figure 5</figcaption></figure><p>Based on the color mapping by neighborhood in figure 5 above, <strong>Williamsburg (in Brooklyn) </strong>was one of<strong> </strong>the top 10 listings from 2015 to 2019, based on the total number of bookings. It is a sort after neighborhood for tourists. And this data shows that the listing price doesn’t need to be outrageous. As shown above, the listing with the highest number of bookings was priced at $60. Placing the 5-year combined listing earnings at a whopping <strong>$5,875,380</strong>.</p><p>However, since earnings is dependent on listing price and number of bookings, it is worth looking at it on both the granular borough to borough level, and a city-wide level per listing. In the Bronx, the highest earning was a $680 listing in Longwood Avenue, totaling <strong>$902,360. </strong>For Staten Island, it was a $700 listing totaling <strong>$766,500</strong>. While in Queens, Brooklyn, and Manhattan, they were $110, $590, and $2,000<strong> </strong>listings respectively, with earnings of <strong>$218,700, $861,400, </strong>and<strong> $2,920,000</strong> earnings, respectively.</p><h3>Conclusion</h3><p>Investing in Airbnb has its risks, as does almost any other form of investment in the world, and during this COVID-19 pandemic, it is more evident than ever. Still, it is worthwhile if you are able to invest without any additional cost i.e. listing your spare room or converting your living room to a second room to rent out, but be advised there are other factors to consider such as ambience, hosting skills, etc.</p><p>So, the question to ask yourself is:<strong> how bad do I really want that extra income, and is it worth the stress of hosting strangers</strong>?</p><p>Thanks for reading!</p><p>Be sure to share your comments and suggestions on markets you would like for me to explore.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8fc9883f3559" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>