João Oliveira Soares

A multivariate methodology to uncover regional disparities: A contribution to improve European Union and governmental decisions

Abstract The aim of this paper is to present a new methodology to classify the levels of socio-economic development of a country’s territory, in order to support regional development policy. This classification is obtained through the use of multivariate statistical methods – factor and cluster analysis, and is based on a wide number of demographic, economic, health, education, employment and culture indicators. The Portuguese continental territory is used as the working example. Results lead to the identification of nine axes of socio-economic characterisation, and the division of the Portuguese territory into four regions with differing degrees of development, reflecting the well-known asymmetry between coastal and inland zones. The ‘socio-economic regions’ uncovered with this methodology allow a much more useful characterisation of the Portuguese territory, for policy making, than does the NUTS-2 classification scheme used by the European Union.

A multicriteria model for portfolio management

The paper presents a new model to support the selection of a portfolio of stocks based on the results of the fieldwork undertaken with fund managers and using direct rating, MACBETH and optimisation techniques. The model consists of defining a benchmark portfolio (in this case, the Dow Jones Eurostoxx50) and scoring its different stocks according to several expected return criteria. Based on this multicriteria value analysis, a procedure is proposed to suggest adjustments to the proportions of the stocks in the portfolio. Finally, the risk of this modified portfolio is taken into consideration in an optimization module that includes constraints concerning the limits of variation for the proportion of each stock.

The European regional policy and the socio-economic diversity of European regions: A multivariate analysis

Abstract There are significant differences among the European Union regions, which have been heightened due to the most recent enlargement in 2004. This paper aims to analyse this diversity and to propose a classification of European regions that is adjusted to the different axes of socio-economic development and, simultaneously, is useful for European regional policy purposes. Multivariate statistical techniques allow the identification of clusters of socio-economic similarity, which are contrasted with the classes considered in the financial proposal of the European Commission (EC) for the period 2007–2013.

Portfolio robustness evaluation: a case study in the electricity sector

AbstractManagers continually face the task of allocating resources to projects when there is not enough money to fund them all. Portfolio Robustness Evaluation (PROBE) is a multicriteria decision support system developed to help managers to perform that difficult task. This paper presents a PROBE model, developed for an electricity distribution company, to select the best portfolio of projects, subject to budget constraints for different types of projects and various organisational units in multiple time periods. Projects requiring large-scale investments are analysed separately from the small-scale projects. The robustness of the selected portfolio of large-scale projects is analysed in an iterative process where broader uncertainty ranges are considered for the values of the projects, and also when an environmental impact criterion is added to the evaluation model.