Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library: stats
See also: draftman factor

Quantlet: pca
Description: PCA performs a Principal Component Analysis for x. It is possible to choose interactively between different criteria for the PCA's and confidence intervals.

Usage: pc = pca (x)
Input:
x n x p matrix
Output:
pc.y n x p matrix principal components
pc.gamma p x p matrix of eigenvectors
pc.lambda p x 1 matrix of eigenvalues

Example:

; loads the library stats

library("stats")   

; loads the library graphic

library("graphic")                                       

; reads the swiss banknote data

x  = read("bank2")             

; shows the principal components of x                          

pc = pca(x)                             

Result:

The graphic is divided into two displays. One shows a 

a scatterplot matrix of X. The second shows at the top

a parallel coordinate plot of gamma (matrix of the 

eigenvectors) and at the bottom the scree plot. 

Interactively you can choose between different criteria

for the number important principal components and 

confidence intervals for the eigenvalues (assumed that

they are really all different).


Library: stats
See also: draftman factor

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Author: Katharina Kroel, Kerstin Zanter, 961202, Sigbert Klinke 970820
(C) MD*TECH Method and Data Technologies, 21.9.2000