Is it possible to predict real estate prices only by indirect parameters? Not by comparing prices with other houses in the neighborhood?
Yes, that’s possible. There are a lot of different factors that can affect the price of a property, such as location, size, age, condition, the ecology of the area, the closeness of schools and highways, the crime rate and so on. If we have all this data, we can determine the price of houses with a high degree of certainty.
However, when it comes to predicting these prices with accuracy, it’s not always a straightforward task. While real estate experts use various methods like comparative market analysis and appraisals to determine the value of a property based on features and area trends, there are many more variables at play than just these. Things like market volatility and economic changes can also have an impact on real estate prices in ways that may be difficult to anticipate or forecast accurately.
Nevertheless, many listings websites and real estate agencies are successful in forecasting real estate prices by neighborhoods and cities. Sometimes it turns out rather crudely, but it all depends on the valuation method.
In this post I will show you how I used different machine learning models to predict house prices in Boston, USA. I know this dataset is very old and has been analyzed thousands of times, but I am sure you will find something new and interesting in my approach.