Reducing information asymmetry in used-car markets by using machine learning models

Information asymmetry in used-car markets results from knowledge differences between buyers and sellers about used cars. Naturally, someone who owns a used car for a certain period, develops a deeper understanding of the real value opposed to someone who did not. The goal of this work is to attempt to reduce information asymmetry in used-car markets by using state-of-the-art machine learning models. With data provided by a Polish used-car online marketplace, a price range estimation as well as a point estimation will be made for every car. A Median Absolute Percentage Error of 7.86%and Target Zone of 58.38% are achieved