How to change probabilities in a cloned tree

Hi, I would like to make a clone that has the same structure as the cloned model but with different transition probabilities. It is possible to do this and if so, how?

0

Comments

3 comments
  • Yes, you can.

    You must reference the same variable names in the clone master and clone copy.

    However, any of those variables can be defined differently for the clone master and clone copy be placing one variable definition just to the left of the clone master and a different variable definition for the same variable name just to the left of the clone copy.

    You can then use those two separate definitions to send different values for the same variable into the clone master and copy.

    To create a variable definition at a specific node, right-click on the node then choose Define Variable from the context menu.

    0
    Comment actions Permalink
  • Hi, I also would like to assign different transition probabilities and payoffs in different clone copies (I have more than 5 clone copies). I am not quite clear on the suggestion mentioned above. Could you provide step-by-step detailed instructions or any tutorials which I can refer to? 

    Thank you

    0
    Comment actions Permalink
  • The same logic applies whether it is a probability, payoff or any other value.

    • Create a variable that is not defined at the root node.
      pProgress
    • Reference that variable in the clone master (and clone copies by default).
    • Create variables for that value that are specific to each clone master/copy and define those at the root node.
      pProgress_Str1 = number
      pProgress_Str2 = number
      etc.
    • At the base of the clone master and copies (note the root node), define the common variable by setting it equal to the variable above for each specific clone master/copy.
      pProgress = pProgress_Str1
      pProgress = pProgress_Str2
      etc.

    This is described in TP2023 R1 Help section 26.4.5.

    1
    Comment actions Permalink

Please sign in to leave a comment.

Didn't find what you were looking for?

New post