1: \begin{abstract}

2:   \noindent {\sc Abstract.} 

3: In this paper, we develp a functional differentiability approach for solving

4: statistical optimal allocation problems.

5: We first derive Hadamard differentiability of the value function through 

6: a detailed analysis of the general properties of the sorting operator. 

7: Central to our framework are the concept of Hausdorff measure and the area and coarea integration formulas from geometric measure theory.

8: Building on our Hadamard differentiability results, we demonstrate how 

9: the functional delta method can be used to directly derive

10: the asymptotic properties of the value function process for binary constrained optimal allocation problems, as well as the two-step ROC curve estimator. 

11: Moreover, leveraging profound insights from geometric functional analysis on 

12: convex and local Lipschitz functionals, we obtain additional generic 

13: Fréchet differentiability results for the value functions of optimal allocation problems.

14: These compelling findings motivate us to study carefully the  first order approximation of the optimal social welfare. 

15: In this paper, we then present a double / debiased estimator for the value functions. 

16: Importantly, the conditions outlined in the Hadamard differentiability section 

17: validate the margin assumption from the statistical classification literature 

18: employing plug-in methods that justifies a faster convergence rate.

19: 

20:   \vspace{15pt}

21:   \noindent {\sc Keywords}:  Optimal treatment allocation, Functional differentiability and delta method, Plug-in classification, ROC curve, Double / debiased machine learning 

22: \end{abstract}

23: