This thesis is centered on the development of a novel variational framework to solve a joint quasi-optimization problem, which consists in finding a common maximizer for a finite family of real-valued functions. The primary objective is to establish the existence of solutions under hypotheses that are significantly weaker than those typically found in the literature. While this problem has been studied in different contexts, our approach provides existence under new, less restrictive concavity assumptions on the family of functions as a whole. This thesis develops a novel theoretical framework based on generalized convexity. We employ recently introduced concepts such as \textbf{solidness, weak upper semicontinuity, and the sub-boundarily constant property on $C$} and novel conditions developed herein, namely \textbf{int-quasiconcavity} for functions and families. The methodology involves constructing a specialized adjusted normal cone operator ($N^A$), adapted from the single-function case, and formulating a generalized quasi-variational inequality (GQVI). We prove that a solution to this GQVI provides a \textbf{sufficient condition} for a solution to the quasi-optimization problem. The main theoretical result is a new existence theorem, derived by proving that our operator possesses the necessary properties (e.g., upper semicontinuity) to guarantee a solution to the GQVI. The power of this result is immediately demonstrated through its primary application: since Kantian equilibrium, introduced by Roemer as an efficient alternative to the often-inefficient Nash equilibrium, can be framed as a special case of the quasi-optimization problem, our theorem provides a new, general existence proof for this cooperative concept under significantly weaker assumptions. The thesis then demonstrates the versatility of the variational method in a different context. We reformulate a classical financial equilibrium model, originally proposed with exogenous prices, as a non-cooperative game where Nash equilibrium prices emerge endogenously. This part of the work includes a computational analysis of the model, an investigation of the \textbf{Price of Anarchy (PoA)}, and the use of a \textbf{Tikhonov regularization} method to ensure the selection of a unique equilibrium. Ultimately, this thesis makes a significant contribution by applying advanced variational analysis to problems in equilibrium theory. The developed frameworks provide a new set of tools for establishing the existence of solutions in joint optimization problems and game-theoretic models, particularly in the context of cooperative behavior.

Variational Approaches in Quasi-Optimization and Game Theory: Applications to Kantian and Financial Nash Equilibria

GIORDANO, Mario Diego Emanuele
2026-11-20

Abstract

This thesis is centered on the development of a novel variational framework to solve a joint quasi-optimization problem, which consists in finding a common maximizer for a finite family of real-valued functions. The primary objective is to establish the existence of solutions under hypotheses that are significantly weaker than those typically found in the literature. While this problem has been studied in different contexts, our approach provides existence under new, less restrictive concavity assumptions on the family of functions as a whole. This thesis develops a novel theoretical framework based on generalized convexity. We employ recently introduced concepts such as \textbf{solidness, weak upper semicontinuity, and the sub-boundarily constant property on $C$} and novel conditions developed herein, namely \textbf{int-quasiconcavity} for functions and families. The methodology involves constructing a specialized adjusted normal cone operator ($N^A$), adapted from the single-function case, and formulating a generalized quasi-variational inequality (GQVI). We prove that a solution to this GQVI provides a \textbf{sufficient condition} for a solution to the quasi-optimization problem. The main theoretical result is a new existence theorem, derived by proving that our operator possesses the necessary properties (e.g., upper semicontinuity) to guarantee a solution to the GQVI. The power of this result is immediately demonstrated through its primary application: since Kantian equilibrium, introduced by Roemer as an efficient alternative to the often-inefficient Nash equilibrium, can be framed as a special case of the quasi-optimization problem, our theorem provides a new, general existence proof for this cooperative concept under significantly weaker assumptions. The thesis then demonstrates the versatility of the variational method in a different context. We reformulate a classical financial equilibrium model, originally proposed with exogenous prices, as a non-cooperative game where Nash equilibrium prices emerge endogenously. This part of the work includes a computational analysis of the model, an investigation of the \textbf{Price of Anarchy (PoA)}, and the use of a \textbf{Tikhonov regularization} method to ensure the selection of a unique equilibrium. Ultimately, this thesis makes a significant contribution by applying advanced variational analysis to problems in equilibrium theory. The developed frameworks provide a new set of tools for establishing the existence of solutions in joint optimization problems and game-theoretic models, particularly in the context of cooperative behavior.
20-nov-2026
Variational analysis; Generalized convexity; Quasi-variational inequalities; Nash equilibrium; Kantian equilibrium; Financial equilibrium; Int-quasiconcavity; Price of Anarchy
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11570/3367024
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact