A comparison between principal component analysis approach and human development index

This paper uses principal component analysis to combine various measures ofhuman development in an optimal fashion to create a development index. Weinterpret this finding as theoretical support for the HDI ranking system as a metricof international human development. Since the simple average of the componentindexes yields rankings roughly equivalent to a more complex multivariatetechnique that selects the weights optimally, this appears to be a case where little islost in the simplistic method, and much is gained in terms straightforwardness.Indeed, while the strength of the HDI appears to lie in its easy comprehension, theweights used therein are consistent with multivariate techniques that generateweights optimally.

کلیدواژه ها:

Human Development Index ، Multivariate data analysis ، principal components