How to create a vector that assembles 30000 sets of parameters at random from the specified ranges

Hello,

I'm trying create a vector that assembles 30000 sets of parameters at random from the specified ranges.

I have a total of 30 parameters and each parameter values will be randomly selected from the specified ranges.

My initial code is something alone the line below;

par1 = 0.01+(0.2-0.01).*rand(30000,1); %when min=0.01 and max=0.2
par2 = 0.1+(0.5-0.1).*rand(30000,1); %when min=0.1 and max=0.5
% repeat this until par30
par = [par1 par2 par30]
  

Instead of doing this, is there a more efficient way to write the code? maybe using loop?

Thank you in advance,

[Adam Danz]

We only have 1 sample of the pattern which is not enough to confirm that we've guessed the pattern correctly.

How are the bounds defined for par1:par30? Is the rule that the low bound keeps getting multiplied by 10? What's the rule of the high bound?

In addition to the bounds, why is par1 adding 0.01 and then par2 is adding 0.1?

Lots if critical details missing.

[Jung]

30 parameter boundaries are defined based on literatures. There is no specific patterns or trends.

Parameter 1 simply has a minimum boundary of 0.01 and a maximum boundary of 0.2 whereas parameter 2 has a min boundary of 0.1 and a maximum boundary of 0.5.

[Adam Danz]

Ah, I see. I thought you wanted to automate the bounds as well. That would be fairly easy to do if there were a pattern.

I suggest making a simple local function where you just need to enter the bounds and the number of random samples you want. Then you just need to call that function for each param. See answer.

Answer 1

[Adam Danz]

You could writing a function that receives the lower and upper bounds as inputs along with the number of values to choose.

function d = boundedRandomNumbers(LB,UB,N)
% Draws N random samples between lower bound LB and upper bound UB
d = rand(N,1)*(UB-LB) + LB;

To get 30k values between the bounds,

par1 = boundedRandomNumbers(0.01, 0.02, 30000);
par2 = boundedRandomNumbers(0.1, 0.5, 30000);

If your bounds were stored in an 30x2 matrix, you could even loop through each row of the matrix and store the parameter values in a 2nd matrix.

% bounds is 30 x 2 matrix of [lower, upper] bounds
parAll = zeros(30,30000); 
for i =1:30
    parAll(i,:) =  boundedRandomNumbers(bounds(i,1), bounds(i,2), 30000);
end