Consider a scenariotree, with 4 stages. The first stage is the root node. This root node has two children. So the second stage has 2 nodes. Each node of the scenario tree has 2 children. So in total we have: 15 nodes. How many nodes will the deterministic equivalent of this have? asked 25 May '14, 17:05 spyimp 
I'm not sure I understand your scenario tree structure. Therefore, I would give a hypothetical example to help you understand the reasoning procedure. Let's assume your problem has three random parameters, each with two possible outcomes (let's say zero or one). First, note that the root node doe not represent any of the random parameters, and each column after the root node represents a random parameter. Consequently, columns #1, #2, and #3 would have 2, 4, and 8 child nodes. Ultimately, the probability space space, \(\Omega\) would have eight scenarios (equal to the number of nodes in the last column). Assuming that each parameters takes either zero or one with equal probability, the probability of each scenario would be \(0.5 \times 0.5 \times 0.5 = 0.125\). $$ \Omega = \begin{Bmatrix} \xi_1=(0, 0, 0),p_1 = 0.125\\ \xi_2=(0, 0, 1),p_2 = 0.125\\ \xi_3=(0, 1, 0),p_3 = 0.125\\ \xi_4=(0, 1, 1),p_4 = 0.125\\ \xi_5=(1, 0, 0),p_5 = 0.125\\ \xi_6=(1, 0, 1),p_6 = 0.125\\ \xi_7=(1, 1, 0),p_7 = 0.125\\ \xi_8=(1, 1, 1),p_8 = 0.125 \end{Bmatrix} $$ answered 26 May '14, 08:17 Ehsan ♦ I totally agree with everything. So we can say ' minimize the expected cost under the uncertainty in ξ' , and we can solve this problem using stochastic programming. So the question is : How would the probability space look for the 'deterministic equivalent' problem ?
(26 May '14, 14:32)
spyimp
Any clues about the deterministic equivalent ?
(27 May '14, 23:06)
spyimp
That's how the probability space looks for the DEP. To write the DEP, you have to replicate each constraint set with randomness for each scenario. Also, each objective function term with randomness would be summed over scenarios. If you don't know how to write the DEP, you might consult a book on SP. For example, the farmer example in Birge and Louveaux is a famous one. It's available in many lecture notes on the web (e.g., see here).
(28 May '14, 04:22)
Ehsan ♦
So the DEP will be as follows:
What do you think ?
(30 May '14, 14:17)
spyimp
(30 May '14, 14:40)
Ehsan ♦

Stage is a reserved term in SP for decisionmaking milestones. I think you mean you have four parameters. Is that correct?
Yes indeed. I do not mean decision milestones. I mean realizations of the uncertain parameter, usually illustrated with a circle.
So my tree has one uncertain parameter. And again, it has 15 circles (nodes). Ie it has 4 epochs. The first epoch has 1 root node. The second epoch has the 2 children of the root node. The third epoch has all the children of the previous epoch. And similarly for the fourth epoch. So the fourth epoch has 8 nodes.
My question is : if I decide to make the deterministic equivalent of this  how many circles (ie nodes) ie realizetions of the uncertain parameter will be ?