Machine Learning with R Cookbook - Second Edition

How to do it...

Perform the following steps to determine the number of principal components with the Kaiser method:

1. First, you can obtain the standard deviation from swiss.pca:

        > swiss.pca$sdev
        Output
        [1] 1.6228065 1.0354873 0.9033447 0.5592765 0.4067472 

2. Next, you can obtain the variance from swiss.pca:

        > swiss.pca$sdev ^ 2
        output
        [1] 2.6335008 1.0722340 0.8160316 0.3127902 0.1654433 

3. Select components with a variance above 1:

        > which(swiss.pca$sdev ^ 2> 1)
        Output
        [1] 1 2

4. You can also use the scree plot to select components with a variance above 1:

        > screeplot(swiss.pca, type="line")
        > abline(h=1, col="red", lty= 3)

Select components ...