Visualizing categorical data

Visualization is an important aspect of any kind of exploratory analysis. Suppose that we had no additional information on the customer's history and we merely know that there were 700 good loans out of thousand sanctioned. In this case, the prediction for the next customer to have a good loan payment would have a probability 0.7. When the variable of interest is a numeric variable, we can plot histograms, boxplots, and so on, to get a good understanding of the problem. Here, the output good_bad is a factor variable and such a plot won't make any sense. A simple understanding of the distribution of the factor variable can be obtained using the bar plot where we would have two bars with the length reflecting the proportionate frequency. However, a bar plot is a one-dimensional plot in the sense that the x-axis labels as good and bad won&apos...