Finding a dream house? Data help!

Posted on Aug 22, 2016

The Portland, Oregon metro area has a lot to offer. A 90 minute drive west brings you to the beach. In the opposite direction there's a snow-capped mountain. It has a thriving food and wine scene, a burgeoning tech sector, and temperate weather. An increasing number of people would like to move to this metro area. Perhaps the greatest concern for these new residents is finding housing. Data can help.

Records for house sales on over the past three years in the neighboring cities of Portland, OR and Vancouver, WA were scraped by using the Python packages BeautifulSoup and Selenium. Around 20,000 sold house records with 15 variables are extracted, including zipcode, address, location, selling price, days on redfin, beds and baths, size, price per sqft, and etc.  The public school ratings were gathered by scraping the website according to zip code. A screenshot is below.

Screen Shot 2016-08-15 at 7.47.17 PM-18

1). When is good time to buy a house?

We see that in the Portland metro area the median house price per square foot increased 35% from Aug 2013 to Aug 2016.  However, the house transaction volume was increasing during the past three years.


The house transaction volume is strongly seasonally dependent - it is highest in summer and lowest in winter. The new school year starts in September, so the real estate market is influenced by people looking to move within a desirable school district. Since offers need one or two months to be processed, the data shown here is delayed. Agreements for house sales usually happen in the Spring. Therefore, this is the hottest season for real estate market.


2). Where are the best locations?

Vancouver, WA and Portland, OR are the twin cities in the metro area. However, which of these is the best place to live?  In general, the living cost in Portland is more expensive than Vancouver as shown in the boxplot. The red boxes show the price of each zip code area in Portland and the green boxes show those in Vancouver. In seven out of ten zip code areas in Portland the median price per unit area is above $200/sqft. The median price per unit in all the areas of Vancouver is lower.


3). Which area has significant appreciation?

During the past three years, the price growth rate per unit area demonstrates the potential appreciation capability. The top 10 zip codes are in Portland and the bottom 10 zip codes are in Vancouver. In this regard, some areas in Portland, such as 97202,97206,97220, have gradually increasing house prices. Buying a house in these areas would conceivably be more rewarding in the long run. In comparison, the price appreciation in Vancouver is relatively weak.


4). Are the public schools good?

The public school ratings for elementary schools, middle schools and high schools were webscraped. The ratings were averaged by zip code since the house data was segmented in this way. The school ratings are positively correlated as shown in the plots.  Especially for elementary and middle school ratings the correlation coefficient is 0.9083869, which is very high. Usually good elementary, middle, and high schools are bound together and vice versa.


As shown in the plots, the housing cost in Portland is higher than that in Vancouver for the same education quality. Also, as school ratings increase, so do house prices.



5). Finally: Picking specific house features in your chosen area

Now you know when and where to narrow your target for house hunting based on price, location, and school district. What specific features do you care about in a house? The following plot shows the correlation relationships among those numerical features. As previously explained, school ratings among elementary, middle, and high schools are highly correlated. The price, house size (sqft), number of bedrooms, and number of bathrooms are positively correlated. Larger houses, more bedrooms, and more bathrooms lead to higher house prices. In general, one would expect smaller houses to be cheaper than larger houses. However, when one takes into consideration the price per square foot, this relationship doesn’t really hold.


As observed, for the house in one zip code and within a short time period (August 2015 to August 2016),  house price is positively correlated with house size (sqft). However, the price per unit square foot shows a complex trend. When the size is less than 2000-sqft, the price per unit square foot drops. It is constant from 2000 to 5000-sqft until it slightly decreases after 5000-sqft. Therefore, buying a house smaller than 2000-sqft means you pay more for each square foot. The combination of house price and house size should be optimized before submitting an offer.



As we see, the analyses of housing historical records are able to help residents find a desired house, fully with trust of data.



About Author

Jingyu Zhang

Jingyu Zhang has Ph. D degree in Electrical Engineering with emphasis on optics and electronics. She had been working as display scientist at Sharp Labs of America for four years. Now she is passionate about data science and...
View all posts by Jingyu Zhang >

Related Articles

Leave a Comment

calistaheritage.Org July 25, 2017
I like the report

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 airbnb Alex Baransky alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus API Application artist aws beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep Bundles California Cancer Research capstone Career Career Day citibike clustering Coding Course Demo Course Report D3.js data Data Analyst data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization Deep Learning Demo Day Discount dplyr employer networking feature engineering Finance Financial Data Science Flask gbm Get Hired ggplot2 googleVis Hadoop higgs boson Hiring hiring partner events Hiring Partners Industry Experts Instructor Blog Instructor Interview Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter lasso regression Lead Data Scienctist Lead Data Scientist leaflet linear regression Logistic Regression machine learning Maps matplotlib Medical Research Meet the team meetup music Networking neural network Neural networks New Courses nlp NYC NYC Data Science nyc data science academy NYC Open Data NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time Portfolio Development prediction Prework Programming PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn Selenium sentiment analysis Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau team TensorFlow Testimonial tf-idf Top Data Science Bootcamp twitter visualization web scraping Weekend Course What to expect word cloud word2vec XGBoost yelp