There is a saying that the value of a house depends on three factors: location, location, location. By interpolating real estate listings, we tried to calculate the value of the location. The results of these calculations are the demonstrated maps. In this section we will shortly explain:

1. Many maps with house prices are already available. Why did we make our own?
2. How to use our maps?
3. Data sources

For a more technical explanation, you can read our dedicated blogpost.

Why did we make our own maps?

There are quite a lot of online maps available that give you an idea about the house prices at a certain location. Besides the online data of the government, several newspapers have an online map, for example De Tijd and De Standaard. If all these are available online, why did we feel we could make a better one?

Most of the maps online are aggregations of house prices per municipality. This is a very straightforward technique, but assumes that there is no variation within a municipality. We don't agree with this statement, we believe that there is lots of variation within a municipality. We illustrate this with a clear example:

We hope you agree that our map gives you a better indication of the house prices in Antwerpen and the surroundings. As you can see on the map, within the municipality Antwerpen (postalcode 2000), there are some very expensive neighborhoods and some cheap ones. These details are lost when aggregating by municipality.

How to use our maps?

We made three kind of maps.

• A map with the median price for a house, expressed in euros.
• A map with the median price per square meters livable area, expressed in euros per square meters.
• Three maps with the median price for an apartment for the largest Belgian cities.

We used the median price instead of the average price, because the distribution of the house prices is skewed to the right. The outliers to the right, castles and other exclusive properties, have a too large effect on the average price. The median is a lot more robust against these outliers.

The difference between the map with the median price and the median price per square meters gives interesting insights. Let's take a closer look at the city of Bruges.

Every Belgian will say that the city center of Bruges is expensive. But is it? The data shows us that Bruges is still very affordable in terms of absolute figures. But the median house in Bruges is a small house, and so the price per square meters is quite high!

Original data

The real estate transactions in Belgium are not made available via open data. So we had to use a proxy. The heatmap is the aggregated result of an analysis we have made on a million real estate online listings from Zimmo.

Based on the work we have done for our clients in the financial sector we can state that there is a correlation - with a very low bias - between transaction prices and real estate listing prices . Using the online listing data has another advantage, namely the availability of descriptive features such as the livable area. The combination of the two is extremely powerful.