A preliminary analysis on SSL trends data:
Tomo Eguchi
I conducted PCA’s on Steller Sea Lion trends data. Here are some preliminary results.
0.011 0.003 0.722 6.275 108.154
(a) PCA on 80’s slope, 90’s slope, and 91 intercept:
Variance explained by each PC: 0.452 0.354 0.193
Cumulative variance: 0.452 0.807 1.000
Eigenvectors:
1 0.513 0.735 -0.443
2 -0.668 0.017 -0.744
3 0.539 -0.678 -0.500
The plot PC 1 vs. PC 2 looks like this:

Looks like there are a few groups of points if I look at from a specific angle: (22, 4, 25), (20, 13), (29, 1, 8, 5, 24, 2, 12), and the rest. What do you think?
(b) Next, I added the longitude and latitude as variables and ran a PCA:
Variance explained by each PC: 0.481 0.224 0.206 0.083 0.006
Cumulative variance: 0.481 0.705 0.911 0.994 1.000
Eigenvectors:
1 0.319 0.212 -0.365 -0.615 0.584
2 -0.638 -0.656 -0.270 -0.145 0.265
3 0.416 -0.561 0.663 -0.203 0.177
4 0.541 -0.458 -0.591 0.129 -0.363
5 -0.161 0.026 0.073 -0.737 -0.652
And the plot of PC 1 vs. PC 2:

Of course, the latitude and longitude play a big role in this analysis. The following is a map of the rookeries:

According to Rencher (1995; p. 433), “the component from a given correlation matrix are not unique to that correlation matrix. … Thus the statement that the first component from a correlation matrix accounts for, say, 90% of the variance is not very meaningful. In general, for p > 2 (# variables), the components from a correlation matrix depend only on the ratios (relative values) of the correlations, not on their actual values, and components of a given correlation matrix will serve for other matrices.”