plotting multiple distributions on one graph

What I'm wanting is a little complicated.

I have an matrix, simphystate, that is (20x1000), or 20 timesteps x 1000 individuals, and the elements can range from 1 to 15

I want to plot the distribution of simphystate over each timestep, but I want to exclude points where the element of an individual at any given timestep is 1 (these individuals are dead)

plotting a histogram with so many distributions just blurs together, so I was thinking I could just plot the line representing the distribution. Something that would look like this :

is this possible?

Answer 1

[Star Strider]

I am not certain what your data are, so I will create some.

A = randn(20, 1000);

A = (A - min(A,[],2)) + 1;
A = A .* 15./max(A,[],2) + randi(9, 20, 1);

for k = 1:size(A,1)
    pd{k} = fitdist(A(k,:).', 'normal');
end

x = linspace(1, 30, 150);

figure
hold on
for k = 1:size(A,1)
    plot(x, pdf(pd{k},x), "DisplayName","Row "+k)
end
grid
legend('Location','best')

You could also use plot3 for this, and separate the individual distributions by row number.

.

[Kitt]

I actually ended up using boxplots as suggested by a peer and just replaced the numbers I didn't want with NaN which worked. But thank you for much for your answer and support!!

[Star Strider]

As always, my pleasure!