Principal Component Analysis
Posted: Thu Aug 01, 2013 7:59 pm
Hi!
I'm a new-comer here and not very familiar with forum's environment.I need to use Principal Component Analysis in my thesis and i face some problems.I import 9 time series and i want to extract PC1 of them.So, I open the series as a group and chose View/Principal Components.(i choose the covariance method as the data are normalized). It should be of the a type like this PC1=aX1+bX2+... where X corresponds to the initially imported variables.The coefficients are the eigenvectors (loadings).Right? Then i want to make this PC1 as a time series so as to regress it with other time series.At this point there is a problem i don't know how to do it. Can you please, help me? Are the steps that i did the right ones? Thank you!
I'm a new-comer here and not very familiar with forum's environment.I need to use Principal Component Analysis in my thesis and i face some problems.I import 9 time series and i want to extract PC1 of them.So, I open the series as a group and chose View/Principal Components.(i choose the covariance method as the data are normalized). It should be of the a type like this PC1=aX1+bX2+... where X corresponds to the initially imported variables.The coefficients are the eigenvectors (loadings).Right? Then i want to make this PC1 as a time series so as to regress it with other time series.At this point there is a problem i don't know how to do it. Can you please, help me? Are the steps that i did the right ones? Thank you!