Search

This paper is concerned with simultaneously regrouping regions and sectors when analyzing the relative sectorial specialization of regions and the relative regional concentration of sectors. An automatic two-mode clustering algorithm is proposed with a view toward a concept of overall localization, corresponding to a discrepancy between an actual two-way contingency table (regions × sectors) and an hypothetical table reflecting independence between regions and sectors. This procedure identifies similar regions (respectively sectors) according to the relative sectorial (respectively regional) structure. This algorithm significantly reduces the size of the original table and obtain an optimal collapsed table with low level of information loss vis-à-vis the degree of overall localization. The properties and results of the algorithm are discussed through two applications, namely Argentina and Brazil.

This paper is concerned with simultaneously regrouping regions and sectors when analyzing the rel- ative sectorial specialization of regions and the relative regional concentration of sectors. An automatic two-mode clustering algorithm is proposed with a view toward a concept of overall localization, cor- responding to a discrepancy between an actual two-way contingency table (regions x sectors) and an hypothetical table reflecting independence between regions and sectors. This procedure identifies similar regions (respectively sectors) according to the relative sectorial (respectively regional) structure. This algorithm significantly reduces the size of the original table and obtain an optimal collapsed table with low level of information loss vis-à-vis the degree of overall localization. The properties and results of the algorithm are discussed through two applications, namely Argentina and Brazil.

This paper develops an automatic grouping procedure of regions and sectors based on a greedy biclustering algorithm, focused simultaneously on their relative regional specialization and relative industrial concentration. Based on an association measure for a contingency table (regions x sectors), this procedure enables to i) significantly reduce the size of the original table and obtain an optimal collapsed table with low level of information loss vis-à-vis the degree of global localization; and ii) identify the homogeneous regions according to the industrial structure in terms of sub- and over- specialization in large two-way contingency tables. The properties and results of the algorithm are discussed through the presentation of three applications, namely Argentina, Brazil and Chile. In particular, an object of discussion, in the case of Brazil, is the result that the number of cells of the original table is reduced by 99% while the lost global localization information is 23%.

This paper develops new statistical and computational methods for the automatic detection of spatial clusters displaying an over- or under- relative specialization spatial pattern. A probability model is used to provide a basis for a space partition into clusters representing homogeneous portions of space as far as the probability of locating a primary unit is concerned. A cluster made of contiguous regions is called an agglomeration. A greedy algorithm detects specialized agglomerations through a model selection criteria. A random permutation test evaluates whether the contiguity property is significant. Finally this algorithm is run on Argentinean data. Evaluating the proposed methodology concludes the paper.

This paper develops new statistical and computational methods for the automatic detection of spatial clusters displaying an over- or under- relative specialization spatial pattern. A proba- bility model provides a space partition into clusters representing homogenous portions of space as far as the probability of locating a primary unit is concerned. A cluster made of contigu- ous regions is called an agglomeration. A greedy algorithm detects specialized agglomerations through a model selection criteria. A random permutation test evaluates whether the contigu- ity property is signicant. Finally this algorithm is run on Argentinean data. Evaluating the proposed methodology concludes the paper.

From data in the form of a two-way contingency table “Regions × Sectors”, the concepts of specialization and concentration, built from the analysis of conditional distributions or profiles, is based on discrepancies among distributions: between profiles and a uniform distribution for absolute concepts; between profiles and the corresponding marginal distribution for the relative concepts; or between the joint distribution and the product of the marginal distributions for the global concept. This paper provides an extensive numerical analysis of measures derived from this approach and from other approaches used in the literature and shows that while the different measures under consideration display rather similar numerical behaviours, differences of ranking call for a particular care when interpreting the numerical results.

The intensity of regional specialization in specific activities, and conversely, the level of industrial concentration in specific locations, has been used as a complementary evidence for the existence and significance of externalities. Additionally, economists have mainly focused the debate on disentangling the sources of specialization and concentration processes according to three vectors: natural advantages, internal, and external scale economies. The arbitrariness of partitions plays a key role in capturing these effects, while the selection of the partition would have to reflect the actual characteristics of the economy. Thus, the identification of spatial boundaries to measure specialization becomes critical, since most likely the model will be adapted to different scales of distance, and be influenced by different types of externalities or economies of agglomeration, which are based on the mechanisms of interaction with particular requirements of spatial proximity. This work is based on the analysis of the spatial aspect of economic specialization supported by the manufacturing industry case. The main objective is to propose, for discrete and continuous space: i) a measure of global specialization; ii) a local disaggregation of the global measure; and iii) a spatial clustering method for the identification of specialized agglomerations.