Principal Component Analysis and its Application to Global Inflation Rates and Per Capita GDP of Different Countries

Kumar, Nishant (2006) Principal Component Analysis and its Application to Global Inflation Rates and Per Capita GDP of Different Countries. Masters thesis, Indian Institute of Science Education and Research Kolkata.

PDF (MS dissertation of Nishant Kumar (06MS23)) Nishant-06MS23.pdf - Submitted Version Restricted to Repository staff only Download (1MB)

Abstract

Principal component Analysis (PCA) is a method of reducing the dimension of the data. In practice, we may have a number of variables which are responsible for the effects, but all may not be of much importance. In PCA, we consider a couple of linear combinations of the variables so that these linear combinations will be sufficient to explain most of the variabilities present in the data. The aim of the transformation is that the first component has as high a variance as possible and the subsequent components have as high a variance with an added constraint that it should be uncorrelated to all the previous principal components. This technique has been applied to the aforementioned data sets and some very interesting results are obtained.

Actions (login required)

View Item