This might actually be an exercise in overfitting. Have look here for a simple introduction: https://kevinbinz.com/tag/classification/
Basically, whenever you have a bunch of data, you can use regression to find a curve that nicely fits the data. However, that does not mean that you identified a causal relationship: you just say how the numbers compared in the past. There is no guarantee that the numbers are going to predict how they will do in the future. (And do you really think that 0.377216 is somehow one of the magical cosmic constants of the housing market?)
In machine learning, you have to separate some of the data from your training set, to use them as a test set. In this way, you have a better chance to find out whether the relationship you discovered is spurious.