How to ignore certain columns in a matrix when calculating a mean across all columns in a matrix?

Hi all,

I have a 100x16 double matrix (attached) which when plotted over one figure looks like this.

Now, I want to calculate the mean across all columns in the matrix so that the output is a 100x 1 double. I can do this by following

load 'data'

figure
plot(S_or_z(:,1:8),'r')
hold on
plot(S_or_z(:,9:end),'b')

% get mean

S_or_z_meancycle = mean(S_or_z, 2);

However, when calcuating the mean, I would like ignore the columns indicated in red.

The rule that I would like to apply is that the columns which range (max-min) exceeds 200 should be ignored when calcuating the mean accross all columns.

Can you help please?

Answer 1

[Davide Masiello]

See if this works

new_S_or_z = S_or_z(:,(max(S_or_z,[],1)-min(S_or_z,[],1)) < 200);

Unrecognized function or variable 'S_or_z'.

plot(new_S_or_z)
mean(new_S_or_z,[],2)

[Tomaszzz]

Thanks Davide,

Instead of :

mean(new_S_or_z,[],2)

Should it be?:

mean(new_S_or_z,2)

Otherwise I get:
Dimension argument must be a positive integer scalar, a vector of unique positive integers, or 'all'.

[Davide Masiello]

Yes, you're right! I got the syntax for "min" and "max" mixed up with the one for "mean", sorry about that.