How to find the eigenvector associated with the highest eigenvalues.

2022 5 29
2 个回答
回答已采纳
更新时间：2022 5 29
4 次查看（30 天）

0 个投票

I ran PCA on a matrix and now I am asked to find the eigenvector associated to the highest eigenvalues.
How can it be done? Is the first eigenvetor in score the one associated with the higest eigenvalues?

请先登录，再回答此问题。

 采纳的回答

Matt J
Matt J 2022-5-29

1 个投票

No, the output of eig() is not sorted, but the output of svd() is.

3 个评论
显示 1更早的评论 隐藏 1更早的评论

Nicola Dall'Osto
Nicola Dall'Osto 2022-5-29
So you are confirming that PCA returns sorted scores (eigenvectros)?
Matt J
Matt J 2022-5-29
编辑：Matt J 2022-5-29
It depends what code you have used to perform the PCA. svd() does return sorted singular values
Nicola Dall'Osto
Nicola Dall'Osto 2022-5-29
I used pca function. It uses SVD algorithm by default.

请先登录，再进行评论。

更多回答（1 个）

Torsten
Torsten 2022-5-29
编辑：Torsten 2022-5-29

0 个投票

Why not just trying whether A*x equals lambda_max*x within a certain tolerance ?

类别

在 帮助中心 和 File Exchange 中查找有关 Linear Algebra 的更多信息

产品

版本

R2022a

标签

Community Treasure Hunt

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

Start Hunting!

Translated by