清空筛选条件
清空筛选条件

Princomp help needed to reduce-change a dimension of the matrix

1 次查看（过去 30 天）
Jonny
Jonny 2011-10-26
I have a matrix X 400x500 (rows x column) and I use
[coeff,score,variance]=princomp(X);
%Calculate explained variance
expvar=100*variance/sum(variance);
figure; pareto(expvar);
xlabel('Number of Principal Components');
ylabel('Explained Variance %');
title('Pareto of Explained variance vs. Principal Component Number');
From this figure,I note that the first ten principal components accounts for more that 70% of the total variance
Now, how can I get my new reduced matrix using coeff,score,variance? Please help me,I'm stuck here Thank you

请先登录，再回答此问题。

回答（1 个）

bym
bym 2011-10-26
Just use the first 10 values of coeff & score,i.e.
score(:,1:10)

类别

在 Help Center 和 File Exchange 中查找有关 Statistics and Machine Learning Toolbox 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