Courant Research Centre: Poverty, Equity and Growth - Discussion Papers 172
This paper reviews various treatments of non-metric variables in Partial Least Squares (PLS) and Principal Component Analysis (PCA) algorithms. The performance of different treatments is compared in the extensive simulation study under several typical data generating processes and recommendations are made. An application of PLS and PCA algorithms with non-metric variables to the generation of a wealth index is considered.
Principal Component Analysis PCA Partial Least Squares PLS non-metric variables simulation wealth index