Real Estate Investment: Buy to Sell or Buy to Rent?

April 30, 2017

(This article was first published on R – NYC Data Science Academy Blog, and kindly contributed to R-bloggers)


Real Estate Investment

If you are looking for an investment to generate supplemental income at low risk, then real estate is a good option to consider. Just like any other form of investment, your decisions need to be guided by your end goal. Buying to sell or “flipping” will require a different mindset as compared to buying to rent for passive income. The latter will require an investigation into the rent to value percentage. My Shiny App provides an exploratory data analysis tool to help facilitate your research.

The Data

For this project I chose Zillow Research as my data source. Zillow has a treasure chest of data median rent and value available in CSV format for public use. This Zestimate data is available on State, Metro, County, City, Zip-code, and Neighborhood levels, for Single Family Homes, as well as Coops and Condos. I picked Single Family Homes for fairly represent the entire country, at the zip-code level as it is a standard geographical unit. Now you might be wondering what a Zestimate is? A Zestimate is Zillow’s estimated rent or value of a particular home. Zillow calculates this number using their proprietary statistical and machine learning models.

I used two datasets for my project, one with median rent, and the second with median value data. I merged the two data-sets to narrow down to the zip-codes for which both rental and value data was available. I also limited the time series to a five year period from January of 2012 to December of 2016 to eliminate missing values.

Exploration and Visualization

Once scrubbed clean, the data were ready for R’s powerful visualization tools. I used the plotly package in my Shiny App to visualize the three variables of interest Value, Rent, and the Percentage of Rent to Value (Rent/Value). The sidebar of the dashboard can be used to make a selection of State, County, or City. Using the zip-codes of my home county of Queens as an example, we can visualize the change in either variable over time.

Link to my Shiny App:

Screen Shot 2017-04-30 at 10.56.57 PM

The Date Range slider can be moved around to adjust the time period of interest, and clicking “All” will show the change over time for all zip-codes of the selected area. Similarly, a bar-graph visualization shows the exact percentage increase in the average median rent and value over the same period of time.

Screen Shot 2017-04-30 at 11.33.59 PM

Jumping out of the graph is Brooklyn (Kings County) with a drastic increase of 42% in rent and 72% in value. Someone looking to “flip” or sell their investment for a profit might want to look at Brooklyn in further depth. The “Map” tab might be useful for a more detailed analysis.

Screen Shot 2017-04-30 at 11.40.19 PM

For this tab, I have used leaflet to display a side by side comparison between either two variables, or the same variable over time. The zip-codes of Northern Brooklyn bordering Queens have seen a positive change over the past five years. Switching gears to rental investment, we can next compare median rent to median value side by side. We see that areas of high value do not necessarily show a proportionately higher rent.


Screen Shot 2017-05-01 at 12.49.07 AM

Sheepshead Bay (√) seems to fare well with a median rent of $2360, with a much lower median value of $778,000. For a rental investment it seems to make sense to consider Sheepshead Bay as compared to a much more expensive area like Boerum Hall (×) with a median rent of about $500 more ($2830) and a median value four times greater ($3.1 million). Looking at this one might conclude that Sheepshead Bay is more bang for the buck.

Screen Shot 2017-05-01 at 12.56.17 AMScreen Shot 2017-05-01 at 12.57.43 AMScreen Shot 2017-05-01 at 12.57.05 AM

Shifting our focus back to Queens, we can see which areas are favorable in terms of value (left), rent (right), and rent to value percentage (center) as a balance between the two. For rental income, it might be a good idea to keep the rent to value percentage in mind while deciding on an investment. Areas of high rent and low value yield the highest rent to value percentage. However, neighborhoods of lower value might not always be ideal for investment as other factors beyond this data need to be taken into consideration.

Screen Shot 2017-05-01 at 1.19.50 AM

Future Directions

More data would definitely be helpful to get a better understanding. Additional data on median income, taxes, schools, and crime would be beneficial to performing some predictive analysis. Also data over a larger time span would paint a broader picture of these trends.








The post Real Estate Investment: Buy to Sell or Buy to Rent? appeared first on NYC Data Science Academy Blog.

To leave a comment for the author, please follow the link and comment on their blog: R – NYC Data Science Academy Blog. offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...

If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Comments are closed.


Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)