We have been given a data set
http://www.yoshizoe-stat.jp/mva/data/carmean2.dat for some reason the following car is missing: volvo 3.8, 2.3, 1.9, 4.2, 3.1, 3.6, 1.6, 2.4.
Stated in Question Information:
The covariance matrix has ranked eigenvalues (5.56, 1.15, 0.37, 0.1, 0.08, 0.05, 0.04, 0.02)' with corresponding vector phi of cumulative proportion of the explained variance as (0.76, 0.91, 0.96, 0.99, 0.99, 1.00, 1.00)'.
Find the principal components.
I have inputed the data into excel and found the covariance matrix and then inputed the covariance matrix into an online matrix eigenvalue/eigenvector calculator (Online Matrix Calculator) and it gives eigenvalues as (5.10064, 1.19045, 0.34573, 0.10859, 0.09077, 0.04971, 0.02245).
So why are these answers different to those stated in the question?
I have inputed all the data accurately and the =COVAR on excel correctly.
Clearly using the eigenvalues i've found will lead to different eigenvectors than if i used the eigenvalues in the question.
Any help would be appreciated.