How to do a boxplot for three samples of different sizes?

Hey everyone,

I am trying to create a boxplot for errors values of three different models. Each error vector has a different size. On the internet I already found a solution for integrating two vectors of different sizes into one boxplot:

 x = randn(600,1);
    y = randn(1440,1);
    z = [x; y];
    g = [zeros(length(x),1); ones(length(y),1)];
    boxplot(z,g);

Is there a way to to this for three vectors??

Answer 1

[Thorsten]

x1 = randn(600,1);
x2 = randn(1440,1);
x3 = randn(500, 1);
x = [x1; x2; x3];
g = [zeros(length(x1), 1); ones(length(x2), 1); 2*ones(length(x3), 1)];
boxplot(x, g)