Continuing with the detection of economic problems in European countries, in this chapter, we will try to replicate country ratings, provided by Standard & Poor's, using both quantitative and qualitative information.
This chapter is an interesting real-case application, because we will use some basic text-mining techniques to replicate Standard & Poor's credit ratings. For this purpose, we will use the country reports issued by the European Commission for the European member states.
We will perform a text-mining process to extract combinations of words or useful terms to predict sovereign ratings.
The following topics will be covered in this chapter:
- Predicting country ratings using macroeconomic information
- Implementing decision trees
- Predicting sovereign ratings using European country reports