Working Paper: Affine Gateaux Differentials and the von Mises Statistical Calculus
Abstract
This paper presents a general theory of one-dimensional differentiability for functionals having convex domains which are not necessarily open. The local approximation is carried out by an affine functions, rather than a linear one as in the standard Gateaux derivative. These derivatives have many applications but their theoretical properties are underexplored in literature. We also analyze variations such the affine counterpart of the classical Hadamard and Frechet derivatives. Applications to Statistics and Economics are also discussed.
Supplementary notes can be added here, including code and math.
Lorenzo Maria Stanca
Assistant Professor of Economics
Greetings! I hold concurrent appointments as an Assistant Professor at Collegio Carlo Alberto and within the Department of Economics, Social Studies, Applied Mathematics and Statistics (ESOMAS) at the University of Turin. My academic focus is centered on economic theory, with a particular emphasis on decision theory.