M. V. Alba Fernández, M. D. Jiménez Gamero
A Digital Elevation Model (DEM) is a bare earth elevation model representing the surface of the Earth. To compare two DEMs, a multivariate and spatialized perspective is proposed. This way, to determine whether two DEMs can be considered as similar: (i) a spatialization (natural or artificial) of the study area is needed, (ii) from each spatial unit, the same categorized variable is derived from each DEM for further comparison, and (iii) the similarity between two DEMs will be equivalent to test a large number of the equalities of categorized variables. A test statistic based on the Euclidean distance between the estimated proportions in each pair of populations is designed. The null distribution of the test statistic is unknown and its asymptotic null distribution is derived. A simulation study is carried out to evaluate the behaviour of the proposal in a wide range of scenarios. Finally, an application to two DEM datasets from the same zone (Allo, Navarra, Spain) is detailed.
Keywords: Euclidean distance, two-sample, multinomial
Scheduled
Invited Session Recent advances in goodness-of-fit and k-sample tests
June 10, 2022 10:10 AM
A13