Statistical and network-based analysis of Italian COVID-19 data: communities detection and temporal evolution

Marianna Milano; Mario Cannataro

We analyzed the trend of each type of data for the period February 24 to March 29, 2020. We computed the main descriptive statistics for all regions in the study period. All analysis are performed by using R software [6] . The Figure 1 conveys the evolution of all dataset over days. As a preliminary test, we applied Pearson's chi-square test. Since p-value was less than 0.05 for each distribution, we decided to use nonparametric test for the following comparison. As initial step, we used the Wilcoxon Sum Rank test to carry out an analysis within the same type of data for all weeks and then, for each single week. The Wilcoxon test is a non parametric test designed to evaluate the difference between two treatments or conditions where the samples are correlated. The Wilcoxon test performs a pair-wise comparison among regions with the goal to evidence which ones show different trend. For this reason, we built a similarity matrix for each couple of regions, for each of the available COVID-19 data. Table 1 reports the similarity matrix related to Hospitalised with Symptoms network in the observation period. We reported the all similarity matrices computed for Italian COVID-19 data in the observation period e in the sigle week in Supplementary file, for the lack of space.