Trust in the European Central Bank: Using Data Science and predictive Machine Learning Algorithms

Abstract

Purpose: This empirical scientific research project aims to apply data science and machine learning tools to determine the influence of different factors on the level of trust in the European Central Bank. This research based on the data from the European Commission’s Eurobarometer Survey 89. The paper also aims to represent some predictive analytics techniques to anticipate the level of confidence towards cental bank. Besides that, we build a couple of data visualizing plots, in order to show the main significant impact on the dependent variable. We created the ECB TrustMap Plot, correlation heatmap matrix and Alluvial diagram. Using this plots, we represented changes in network structure over people responses and decision making. Methodology: to calculate the index of trust in the central bank we used Logistic Regressioin, Decision Tree, Random Forrest and Neural Network models. Verify the output and results by using the VIF of the Logistic Model, Cross-validation, Confusion matrix, ROC-curves and accuracy estimations. Main Findings: trust in one-single currency, inflation problems, expectations about the future of EU, indicator of happiness and other indicators has a significant impact on the the level of trust in the central bank.