Gini-type correlation coefficients have become increasingly important in a variety of research areas, including economics, insurance and finance, where modelling with heavy-tailed distributions is of pivotal importance. In such situations, naturally, the classical Pearson correlation coefficient is of little use. On the other hand, it has been observed that when light-tailed situations are of interest, and hence when both the Gini-type and Pearson correlation coefficients are well defined and finite, these coefficients are related and sometimes even coincide. In general, understanding how these correlation coefficients are related has been an illusive task. In this paper, we put forward arguments that establish such a connection via certain regression-type equations. This, in turn, allows us to introduce a Gini-type weighted insurance pricing model that works in heavy-tailed situations and thus provides a natural alternative to the classical capital asset pricing model. We illustrate our theoretical considerations using several bivariate distributions, such as elliptical and those with heavy-tailed Pareto margins.

This paper covers a number of different topics related to the measurement and modelling of dependency within economic capital models. The scope of the paper is relatively wide. We address in some detail the different approaches to modelling dependencies ranging from the more common variance-covariance matrix approach, to the consideration of the use of copulas and the more sophisticated causal models that feature feedback loops and other systems design ideas.

There are many data and model uncertainties in modelling dependency and so we have also endeavoured to cover topics such as spurious relationships and wrong-way risk to highlight some of the uncertainties involved.

With the advent of the internal model approval process under Solvency II, senior management needs to have a greater understanding of dependency methodology. We have devoted a section of this paper to a discussion of possible different ways to communicate the results of modelling to the board, senior management and other interested parties within an insurance company.

We have endeavoured throughout this paper to include as many numerical examples as possible to help in the understanding of the key points, including our discussion of model parameterisation and the communication to an insurance executive of the impact of dependency on economic capital modelling results.

The economic capital model can be seen as a combination of two key components: the marginal risk distribution of each risk and the aggregation methodology which combines these into a single aggregate distribution or capital number. This paper is concerned with the aggregation part, the methods and assumptions employed and the issues arising, and not the determination of the marginal risk distributions which is equally of importance and in many cases equally as complex.

Food Based Dietary Guidelines (FBDG), in order to be attainable, should take into account the cultural context of the population for which they are developed. They need to focus on foods that actually determine the intake of nutrients for which desirable changes have been identified. Inter-individual variability of nutrient intakes – a measure of the heterogeneity of intakes – is determined by the variability in the consumption of its food sources among the population. The foods that determine a high proportion of inter-individual variability in nutrient intake can be identified on the basis of data banks that describe food and nutrient intakes in the population. The experience accumulated to design Food Frequency Questionnaire through selection of key foods suggests that high quantitative contributors to a specific nutrient may not be important determinants of its intake. On the other hand, the Pearson correlation between the intake of each food source and the total nutrient intake allows quantification of the percentage of variability explained by each item and takes into account the possibility of correlations between different food sources. Once a key food is identified, several strategies are available to modify its intake in the population: through changes in the percentage of consumers/in the mean portion size/in the frequency of intake. The anticipated level of change can be predicted according to the strategy adopted.