The identification of similarities and differences between countries is crucial for detecting imbalances and taking corrective measures to avoid the propagation of financial crises. SOM is an unsupervised neural network that is very useful for exploratory data analysis (EDA). The unsupervised nature of SOM is due to the fact that the network is able to learn patterns in the data without using a target variable. Therefore, the different neurons contained in the network self-organize themselves using the input data. The following diagram is the most common representation of a SOM:
As we can see in the preceding ...
