However, M estimators provide for the solutions of the location parameters estimation problem [21 23].
If solutions are acceptable for the decision-maker based on the values of these parameters, the algorithm will stop; otherwise, other solutions are searched by changing these parameters. .
If solutions are acceptable for the decision-maker based on the values of these parameters, the algorithm will stop; otherwise, other solutions are searched by changing these parameters.
Notice, however, that the solution method applies also for arbitrary values of these parameters.
Lemma 3 If k ⁎ ≥ Γ, then the optimal solution consists of the solutions of a robust knapsack problem with parameter Γ for N 1 and of an instance of (KP ) for N 2 with nominal weights.
□ If k ⁎ ≥ Γ, then the optimal solution consists of the solutions of a robust knapsack problem with parameter Γ for N1 and of an instance of (KP) for N2 with nominal weights.
Otherwise, if k ⁎ < Γ, then the optimal solution consists of the solutions of an instance of (E-kKP) with parameter k⁎ and increased weights w ^ j for N1 and a robust knapsack problem with parameter Γ − k ⁎ for N2.
Cross-checking multiple solutions for consistency of the extracted parameters validates the solutions of ρ(r).
Although this may be relaxed to a finite set of solutions in the special case of "flip flop" PK. the process of assessing for a unique solution for the parameters is encompassed within the framework of identifiability analysis.
Each curve represents, for different solutions, the maximum value reached during one period of the solution at different parameter values.
For redundant arms, however, the solution consists of a one parameter set.
