What are the errors in Ben-Haims argument that Info-Gaps robust model is not a Maximin model?
Answer-20: Ben-Haim keeps changing his mistaken explanation as to why Info-Gap’s robustness model is not a Maximin model. So, you may have encountered various versions of the explanation. Be that as it may, the error in the various explanations is due to a profound misconception as to what a Maximin model is. Maximin models are based on the idea that decisions are evaluated on the basis of their worst-case performance with respect to the unknown state of Nature. So translating this into the Info-Gap context, the question that needs to be asked is: what is the meaning of “worst case” in the framework of Info-Gap’s robustness model? Since Info-Gap’s robustness model seeks robustness with regard to a constraint, namely a performance requirement such as r(d,u) ≥ r*, ∀u∈U(α,û), the notion worst case must also be defined with respect this constraint. Yet, for some odd reason Ben-Haim considers a “worst case” to be with regard to the (smallest) value of r(d,u), and not with regard to the cons
