Predicting Housing Prices: The Case of Ankara Housing Market

Abstract

The housing market is crucial for economies, especially in Türkiye, where economic instability has caused rising housing prices. This study aims to predict trends in the city’s rental market using a dataset of 20,000 listings. By using advanced data analysis and machine learning, the research seeks to identify patterns and forecast rental price movements, benefiting households and policymakers. Valuing housing is complex due to its varied characteristics, prompting the use of hedonic pricing theory. This involves a regression model that links housing prices to their features. While traditional linear regression has been used, non-linear machine learning models often provide more accurate predictions. The research offers a framework for predicting housing market trends and fostering equitable urban development, addressing housing needs in Ankara and beyond.