Today, on the Swedish house market there exist many apartments that go on sale each day.
Because people generally want to get a better grip on the final price of apartments, there is
a need for an analytic tool that can empower individuals to make an informed decision when
selling or buying. In this thesis, we present a model that can be used as a complimentary tool
to predict the final price of apartments on Södermalm, Stockholm. The data that the model
is based on is gathered from the broker firm Södermäklarna. The underlying model depends
on variables such as living space, number of rooms, area, among other important factors.
The result shows that almost all of these aforementioned variables are statistically significant
in the model. Furthermore we show that the standard error of the entire regression model is
about 12% and that two of the most important factors are living space and number of rooms.
We also show that in some cases, by just adding a room to an apartment, it can raise the final
price substantially. We conclude in this thesis that the model is capable of predicting prices
on apartments to some extent. But considering the standard error we conclude that there is
room for improvement, one way of doing this is to add more observation objects and variables.
2011. , 48 p.