Size of coeff in PCA example is not correct

Following the demo code for pca exactly gives a different coeff result.

Expected result:

My result:

I have no idea what could be causing this... Any ideas?

Answer 1

[Raunak Gupta]

Hi,

The vector you are seeing in your result is the component wise variance of each principal component. I tried the same example and coeff variable is a 4x4 matrix for me.

Can you try with full syntax which is mentioned below:

[coeff,score,latent] = pca(ingredients) 

This should give correct result.