Reduction of Information Asymmetry in the Used Car Market Using the Random Forest Method
Milosz Bies, Wiesława Gryncewicz, Agata Kozina, Marcin Hernes, Artur Rot, Ryszard Zygała · 2021
Information asymmetry is one of the most common market failures observed in the free market economy. Its various detrimental effects on the economy are especially visible in the used car market. This piece of research aims to develop a tool that by using machine learning could reduce information asymmetry on the used car market. This solution allows for limiting the negative effects of its occurrence. The analysis conducted led to the construction of a tool that enables the reduction of information asymmetry in the used car market. The discussion on the quality of the tool emphasised the importance of exploratory data analysis in the whole process. Also, the main elements that determine the predictive capacity of the random forest model were discussed and the main areas for its further improvement were identified. The interdisciplinary nature of the research was also highlighted, along with the key importance of the domain knowledge in the whole process of analysing data and constructing models.