This dissertation proposes an integrative framework—Efficiency, Synergy, and Returns (ESR) to examine how artificial intelligence (AI) and blockchain technologies contribute to different dimensions of performance in multi-market energy systems. The framework is structured in three phases: (1) AI-based predictive efficiency, where machine learning models improve forecasting accuracy and market responsiveness; (2) blockchain-based coordination, which facilitates interoperability, transparency, and trust across electricity, carbon, and green certificate markets; and (3) value realisation at the market level, which captures how variations in informational and institutional conditions may be associated with firm-level and macroeconomic outcomes. Empirically, the study adopts a multi-method and multi-dataset research design, integrating behavioural finance insights, robust machine learning techniques (including IQR-RANSAC), and institutional analysis. The empirical components operate at complementary technological, institutional, and financial levels, rather than forming a single unified empirical pipeline. Evidence from corporate green innovation under capital market liberalisation shocks suggests that improvements in informational and coordinative efficiency may be associated with differences in innovation outcomes and resource allocation patterns. The findings highlight that AI and blockchain should not be viewed as isolated technologies, but as complementary mechanisms operating across different analytical layers. The ESR framework contributes to the literature by linking micro-level prediction, meso-level coordination, and macro-level value realisation, offering an integrative conceptual perspective alongside practical implications for energy trading, digital market infrastructure, and sustainable finance.
Questa tesi propone un quadro integrato—Efficienza, Sinergia e Rendimenti (ESR) per analizzare come le tecnologie di intelligenza artificiale (AI) e blockchain contribuiscano a diverse dimensioni delle prestazioni nei sistemi energetici multi-mercato. Il framework è articolato in tre fasi: (1) efficienza predittiva basata su AI, in cui modelli di machine learning migliorano l’accuratezza delle previsioni e la reattività del mercato; (2) coordinamento basato su blockchain, che facilita interoperabilità, trasparenza e fiducia tra i mercati dell’energia elettrica, del carbonio e dei certificati verdi; e (3) realizzazione del valore a livello di mercato, che cattura come variazioni nelle condizioni informative e istituzionali possano essere associate a risultati a livello d’impresa e macroeconomico. Dal punto di vista empirico, lo studio adotta un approccio multi-metodo e multi-dataset, integrando elementi di finanza comportamentale, tecniche robuste di machine learning (incluso IQR-RANSAC) e analisi istituzionale. Le componenti empiriche operano su livelli tecnologici, istituzionali e finanziari complementari, piuttosto che costituire un unico percorso empirico integrato. Evidenze relative all’innovazione verde delle imprese in presenza di shock di liberalizzazione dei mercati dei capitali suggeriscono che miglioramenti nell’efficienza informativa e nel coordinamento istituzionale possano essere associati a differenze negli esiti innovativi e nei modelli di allocazione delle risorse. I risultati evidenziano che l’AI e la blockchain non dovrebbero essere considerate tecnologie isolate, ma meccanismi complementari che operano su diversi livelli analitici. Il framework ESR contribuisce alla letteratura collegando previsione a livello micro, coordinamento a livello meso e creazione di valore a livello macro, offrendo una prospettiva concettuale integrata insieme a implicazioni pratiche per il trading energetico, le infrastrutture di mercato digitali e la finanza sostenibile.
Efficiency, synergy, and returns: an integrative ESR framework for AI-driven prediction and blockchain-enabled coordination in multi-market energy systems, with empirical evidence from corporate green innovation under capital market liberalization shocks
FU, JIACHENG
2026
Abstract
This dissertation proposes an integrative framework—Efficiency, Synergy, and Returns (ESR) to examine how artificial intelligence (AI) and blockchain technologies contribute to different dimensions of performance in multi-market energy systems. The framework is structured in three phases: (1) AI-based predictive efficiency, where machine learning models improve forecasting accuracy and market responsiveness; (2) blockchain-based coordination, which facilitates interoperability, transparency, and trust across electricity, carbon, and green certificate markets; and (3) value realisation at the market level, which captures how variations in informational and institutional conditions may be associated with firm-level and macroeconomic outcomes. Empirically, the study adopts a multi-method and multi-dataset research design, integrating behavioural finance insights, robust machine learning techniques (including IQR-RANSAC), and institutional analysis. The empirical components operate at complementary technological, institutional, and financial levels, rather than forming a single unified empirical pipeline. Evidence from corporate green innovation under capital market liberalisation shocks suggests that improvements in informational and coordinative efficiency may be associated with differences in innovation outcomes and resource allocation patterns. The findings highlight that AI and blockchain should not be viewed as isolated technologies, but as complementary mechanisms operating across different analytical layers. The ESR framework contributes to the literature by linking micro-level prediction, meso-level coordination, and macro-level value realisation, offering an integrative conceptual perspective alongside practical implications for energy trading, digital market infrastructure, and sustainable finance.| File | Dimensione | Formato | |
|---|---|---|---|
|
Efficiency, Synergy, and Returns.pdf
non disponibili
Licenza:
Tutti i diritti riservati
Dimensione
4.29 MB
Formato
Adobe PDF
|
4.29 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/382761
URN:NBN:IT:POLIMI-382761