Scopul nostru este sprijinirea şi promovarea cercetării ştiinţifice şi facilitarea comunicării între cercetătorii români din întreaga lume.
Autori: Alexandru I. Petrisor, J. Wanzer Drane, Liviu Dragomirescu
Editorial: University Politehnica of Bucharest Scientific Bulletin Series A: Applied Mathematics and Physics, 68(2), p.19-34, 2006.
Even though its use has been limited, the DAC statistic, defined as the difference between the empirical distribution for the cases and that of the total sample, presents a great potential in detecting spatial clusters. Our previous work focused on examining its sensitivity to the location of origin and orientation of axes, and on its possible role as an instrument to detect clusters. Results indicate that the DAC statistic does not depend on the location of the origin, but is influenced by the orientation of axes. The DAC statistic cannot be used alone to detect clusters, but in conjunction with our methods. GIS-based methods were examined as a possible candidate. Results indicate that results obtained using density maps depend on choices of the users, and kriging approaches are preferable in terms of repeatability. This article summarizes our previous results, most of which had not been published before, and attempts to build up a methodology using the DAC statistic in conjunction with other spatial techniques to detect spatial clusters.
Cuvinte cheie: test DAC, statistici spatiale (bidimensionale), clustere, distributie cumulativa empirica, kriging, geocodificare, harti de densitate, subponderalitate, date din certificatele de nastere // DAC statistic, spatial statistics, clusters, empirical cumulative distribution, kriging, geocoding, density maps, low birth weight, birth certificates data