ABSTRACT
After considering mixtures where the component densities parameterize location, scale, concentration, skewness, and index (Chapter 6), the next step in increasing modelling flexibility is allowing multiple scaled components. An argument as to why this might be desirable is best made by looking at a specific example of data for which multiple scaled components are needed. Consider the bankruptcy data from the MixGHD package. These data contain the ratio of retained earnings to total assets as well as the ratio of earnings before interest and taxes to total assets for 66 American firms (Figure 7.1).
