We want to analyse loans in order to predict likelihood of Default.
To do this, we will experiment with:
- Decision trees
- Random forests
- Neural networks
Each of these models relies on us pre-defining some input parameters. We must assess how well these models perform (depending on the input parameters), in order to find the one that best predicts Default.
The data, and a starting R script for our analysis can be found here.
You must create a copy of the template notebook, and use your copy to try out the methods.
Notebooks are a great way for organising text and code in order to follow through data exploration and analysis.
You can preview the analysis that we will carry out here in Analysis_Template.ipynb.