A non-parametric distance-based method using all available indicators for integrated environmental assessment - a case study of the Mid-Atlantic Region, USA

摘要

The paper presents a multivariate measure useful for integrated environmental assessments. It is a weighted distance measure applied to metric data but based on nonparametric statistical procedures. The proposed measure allows all environmental indicators to be used directly without any reduction in dimension (e.g. as in principal component analysis) nor losing variance while being able to tolerate possible non-normality of the indicators, as well as non-linear relationships among them. Results of the hypothetical example and the Mid-Atlantic case study show that the proposed measure is suitable and valuable for integrating multiple indicators into a single index, an important task in integrated environmental assessment.