A Compositional Statistical Analysis of Capital per Worker

Most of economic literature has presented its analysis under the assumption of homogeneous capital stock composition. However, capital composition differs across countries. What has been the pattern of capital composition associated with World economies? I make an exploratory statistical analysis based on the Aitchinson logratio transformations and the related tools for visualizing and measuring statistical estimators of association among the components. As initial findings could be cited that: 1. It is observed a clear correlation in terms of capital stock participation between two building-industry- related components, 2. Manufacturing behaves differently, especially durable goods sector. 3. There are differences among subsamples.