Interpreting outputs of pca function

2 次查看(过去 30 天)
Hi all,
I am having some trouble with interpreting the outputs of the function "pca" in matlab. If I have a 10x15 matrix, with each row of the matrix corresponding to an observation, and each column corresponding to a variable, and I use that matrix as an input into the "pca" function, I see that the first two principal components explain ~90% of the variance in the data set by looking at the output variable "explained". My question is: Which variables in the original input matrix do these two principle components correspond to? Is there any way of knowing? In other words, which column of variables explains the majority of the variance in the data set? Thank you.

回答(1 个)

Sagar
Sagar 2015-8-9
In short, you cannot attribute one principal component to a particular variable, because by definition, a principal component is a linear combination of all the variables with different weights. But you can determine which variable contributes the most in a principal component by calculating correlation between each variables and the principal component of interest.

类别

Help CenterFile Exchange 中查找有关 Dimensionality Reduction and Feature Extraction 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by