EFFECTIVE DIMENSIONALITY OF ENVIRONMENTAL INDICATORS - A PRINCIPAL COMPONENT ANALYSIS WITH BOOTSTRAP CONFIDENCE-INTERVALS

In this paper, a principal component analysis (PCA) is performed on 14 selected environmental indicators with 'bootstrapped' confidence inte rvals. The term 'bootstrap' refers to the process of randomly re-sampl ing the original sample set to generate new data sets and using these new data sets to make estimates of the statistic of interest The objec tive is to derive some quasi-confidence intervals for the statistics w hen the underlying statistical distributions of the statistics are unk nown. The analysis indicates that the first four principal components, which together account for more than 60% of the total variance in the original 14 variables, appear to be statistically significant based o n the bootstrapped eigenvalue method, although the bootstrapped eigenv ector method seems to be more conservative by identifying only the fir st two components as the significant ones. The first four principal co mponents have large coefficients (eigenvectors) in absolute values wit h ail; biodiversity, land and wafer indicators, respectively All these suggest that there is large redundancy in the existing environmental indicators. Consequently, to avoid overwhelming and confusing indicato r-users including decision makers and the general public, developing f our sub-indices representing air, water land and biodiversity should b e the primary focus, which would probably capture the most important a spects of the environment. (C) 1998 Academic Press.