Principal Component Analysis to create Indicator

[jamiesh]

I am using a data set containing 41 monthly variables (exports, imports, industrial production, etc.) from five different countries and performing Principle Component Analysis. The first principal component explains more than 75% of the variance of all variables. My goal is to now use the first principal component to create and indicator that tracks gdp for the region.

I have utilized the eigenvectors (loadings) as weights and multiplied each variable with the corresponding weight and added them all up to create a series for the first principal component. I have calculated the y/y % change however it does not match up with the actual regional GDP data. I am wondering if anyone can guide me as I think I might be doing something wrong. Any help would be greatly appreciated.