Sampling iteratively random value from a matrix/array

k = unique(ID_site)
rl = [0.7 0.8 0.85 0.95 1]
rs = [0.65 0.7 0.75 0.8 0.85 0.9 0.95]
im = [0.55 0.65 0.75 0.85 0.9 0.95 1]
ka = [0.92 1]

for i=1:numel(k)
tc1 = rl.*rs.*im.*ka
tc2 = rl.*rs.*im.*ka
tc3 = rl.*rs.*im.*ka
end

Hello,

I want to calculate the products tc1, tc2 and tc3 several times (10000 nsamples) for each ID_site, sampling each time a random value from rl, rs, im and ka.

The output should display 10000 results of tc1, tc2 and tc3 for each test site.

Does anyone knows how to do that?

Thank you very much

Answer 1

[Stephen23]

No loops required, just use indexing:

k = 1:13; % fake data
n = numel(k);
rl = [0.7 0.8 0.85 0.95 1]
rs = [0.65 0.7 0.75 0.8 0.85 0.9 0.95]
im = [0.55 0.65 0.75 0.85 0.9 0.95 1]
ka = [0.92 1]
% random indices:
rl_x = randi(numel(rl),1,n);
rs_x = randi(numel(rs),1,n);
im_x = randi(numel(im),1,n);
ka_x = randi(numel(ka),1,n);
% output:
out = rl(rl_x) .* rs(rs_x) .* im(im_x) .* ka(ka_x)

[Enrico D'Addario]

Thank you Stephen

but I need to calculate "out" for 1000 times for each "k".

I have e.g. 110 id_sites, for each id_site I need 1000 values of "out" .

I think the code should be like:

for each id_site, calculate 1000 values of out, sampling randonmly values from the variables rl, rs, im and ka

Am I wrong?

[Stephen23]

"Am I wrong?"

So far nothing you have described requires a loop. You can trivially generate the random indices with size 1000x110 and then get all of your results at once.

https://www.mathworks.com/help/matlab/matlab_prog/vectorization.html

s = 1000; % number of samples
k = 1:110; % fake data
n = numel(k);
rl = [0.7,0.8,0.85,0.95,1];
rs = [0.65,0.7,0.75,0.8,0.85,0.9,0.95];
im = [0.55,0.65,0.75,0.85,0.9,0.95,1];
ka = [0.92,1];
% random indices:
rl_x = randi(numel(rl),s,n);
rs_x = randi(numel(rs),s,n);
im_x = randi(numel(im),s,n);
ka_x = randi(numel(ka),s,n);
% output:
out = rl(rl_x) .* rs(rs_x) .* im(im_x) .* ka(ka_x);

[Enrico D'Addario]

Thank you so much!