Analysing the Determinations of Financial Inclusion in Africa Based on Random Forest Model and Logistic Regression Model

Abstract

The goal of this research is to look into the factors that influence financial inclusion in Africa. This research uses logistic regression method and random forest method to find the influence and importance of different determinants of financial inclusion. Besides, the researcher compares the estimation accuracy of these two models by evaluation matrix and cross-validation score. The result shows that logistic regression model is better than random forest model when estimating the financial inclusion among African people. In addition, it can be found that age, phone ownership, and household size have significant effects on bank account opening. The analysis of this topic is important for the improvement of living standard of African people.