Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Factor scores

By Patrick | Jan 26, 2017 03:49PM CET

Hi, Thanks for a great tool! I am performing a principal component analysis. The tutorial on this topic gives no explanation, as far as I can see, on the table 'Factor scores'. I was under the impression that those numbers were obtained by multiplying t[factor loadings] with (X-X') where X is a vector of original variables and X' is a vector of means for each variable. If I do this, I end up with another results than your tool. Could you explain what the tool does in order to compute the Factor Scores? Thank you in advance!




By Sébastien | Jan 27, 2017 06:55PM CET | XLSTAT Agent

Hi Patrick

In a standard PCA built on a Pearson (n) correlation matrix, in order o obtain a factor score, say for the first obsertation on F1:
- You multiply the observation’s standardized data by the variables’ eigenvectors on F1.
- You sum up the results and voilà!

Jean Paul

Post Your Public Answer

Contact Us
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
Invalid characters found