Stochastic Programming: pre-sampled discrete random VECTOR

Problems with modeling
leo
User
User
Posts: 1
Joined: 3 months ago

Stochastic Programming: pre-sampled discrete random VECTOR

Postby leo » 3 months ago

Good morning,

Thank you in advance for your help.

My problem is related to the definition of a random vector of variables for each scenario in a 2stage stochastic programming problem using EMP.

I have the pre-sampled discrete distributions for equiprobable scenarios: each scenario has a predefined vector of random variables.

For example: I have 4 scenario each one with probability 1/4. For each scenario I want to define a vector of pre-sampled random variables (reading values from a file).
scenario 1 random vector d = [10 20 30]
scenario 1 random vector d = [20 25 10]
scenario 1 random vector d = [10 17 22]
scenario 1 random vector d = [5 10 15]

Just to be clear, I want to have just 4 scenarios (in this example) and I do not want to associate a probability to each element of the random vectors.

How can I do that?
N.B. I do not want to define a random factor for each scenario that influences my random variable.

I cannot find a solution or a previous discussion related to this problem.
Thank you for your help.
Egidio

Lutz
User
User
Posts: 12
Joined: 1 year ago

Re: Stochastic Programming: pre-sampled discrete random VECTOR

Postby Lutz » 2 months ago

Hi,

You need to use the emp keyword "jrandvar". Here is an example where this is used:

https://www.gams.com/latest/emplib_ml/l ... joint.html

More general information can be found here:

https://www.gams.com/24.9/docs/UG_EMP_SP.html

I hope that helps,
Lutz


Who is online

Users browsing this forum: No registered users and 1 guest