How much is the plot?

Inspiration

What is the price of an additional hospital or subway station in my neighborhood? And is that price within or outside my willingness to pay for that amenity? Am I willing to give up an amenity in exchange for a lower land price? Households face these decisions when they want to invest in a property or move. However, prices for amenities and features are often unknown, making it difficult to optimally weigh different amenities against each other. Our tool aims to help with this problem by making the importance of amenities visible using detailed geodata on amenities and land prices for the case of Frankfurt am Main.

What it does

First, our tool allows us to select features and display neighborhoods and land prices that meet our preferences. Our tool also allows us to study counterfactual scenarios for a neighborhood and examine the impact of changing amenities and neighborhood characteristics on land value. For a household’s investment decision, our tool thus allows to compare their subjective willingness to pay for an amenity with its shadow price to trade off different neighborhood amenities optimally. Further, for public planners, our tool allows to understand the elasticity of land prices with respect to factors such as the availability of health care facility and connectedness to public transit to evaluate how urban policies relate to land prices.

How we built it

We draw on open source data that were shared with us, such as BORIS and Census data, but also retrieved open source data on our own to create additional features that influence land values. Our data capture variety of features, such as land use, population and building density, and the demographic composition of a neighborhood. We further retrieve detailed open source data to examine factors such as the composition of the urban area in terms of built-up areas and vegetation, distance to major urban centers, shopping facilities, public transportation hubs, health and education facilities, and noise and air pollution associated with Frankfurt Airport.

Challenges we ran into

No pain, no gain

Accomplishments that we're proud of

Creating a number of features using non-conventional and self-retrieved data on urban and demographic make up and amenities of the neighborhood using geopandas and QGIS
Feeding features into prediction task using Sci-kit learn
Creating interactive visualization of our results in Streamlit