Exploratory spatial analysis for interval data: A new autocorrelation index with COVID-19 and rent price applications

This paper aims to identify the behavior of interval data associated to its respective geospatial information with in the framework of Symbolic Data Analysis. The main idea is to extend Moran’s autocorrelation index of Exploratory Spatial Analysis to interval data. Symbolic data analysis is a domain of research and application related to the areas of machine learning and statistics that provide tools to describe units (objects), enabling them to consider variability. Spatially correlated data are geospatial data with spatial autocorrelation, and the variability that comes from each region and neighborhood may be better expressed by intervals. Thus, this paper demonstrates the importance of considering the variability present in the interval variable and the variability present in geographical information. Experiments with synthetic interval data are performed to illustrate the usefulness of the proposed approach. We also, analyze two applications, dealing with COVID-19 and rent price interval data.