L’autocorrelazione spaziale presente in stime territoriali alternative della dotazione infrastrutturale

The variables on infrastructure endowment are often used in regional growth analysis to explain regional development. In the literature, there are different methods to quantify public capital. The paper analyses four different indexes, built up on the base of the past investments or on the infrastructure endowment data. In both cases the indexes represent spatial data and therefore we analyze the spatial autocorrelation of each index. The aim is to explore spatial structure in the data. We estimate the global autocorrelation index, defined by Moran, that uses all the data for the entire area to investigate the presence of spatial patterns. After we evaluate local statistics to quantify the extent of dependency and inhomogeneity that are included in the data.The results of the spatial autocorrelation analysis point out significative and remarkable spatial patterns in the distribution of infrastructure endowment only if we consider index expressed in physical terms.