Geoscience Reference
In-Depth Information
rng(0)
s1 = 10*rand(30,1);
s2 = 7*rand(30,1);
s3 = 12*rand(30,1);
We use the same mixing equation as in the PCA example to create the three
columns of
x
(corresponding to the three minerals) by linearly mixing the
source rocks
s1
,
s2
and
s3
x(:,1) = 15.4+ 7.2*s1+10.5*s2+2.5*s3;
x(:,2) = 124.0-8.73*s1+ 0.1*s2+2.6*s3;
x(:,3) = 100.0+5.25*s1- 6.5*s2+3.5*s3;
x = x + 3.8*randn(size(x));
save sediments_2.txt x -ascii
and save the data in the i le
sediments_2.txt
(Fig. 9.4). Having created the
synthetic data set, the data stored in
sediment_2.txt
are loaded by typing
clear
x = load('sediments_1.txt');
We can display the histograms of the data to see that they are not Gaussian
distributed.
Fig. 9.4
Sample input for the independent component analysis. h e relative proportions of
three minerals in 30 sediment samples rel ect temporal variations in the contributions from
three source rocks within the catchment of a sedimentary basin.