In complex systems, the aggregate dynamics is generated by the interaction among their individual constituents and it exhibits properties that cannot be inferred by observing attributes of each of them. However, since the features of those individuals induce their decisions, thus affecting their consequent interactions, although it is impossible to establish the causation line between single choices and the aggregate properties of the system, there is room for an investigation on the variation induced on the emergent dynamics by changes in the characterization of individuals. Thus, by means of agent-based models, it is possible to investigate how emergent phenomena of macroeconomic dynamics change in response to modification in the statistical distributions of individual traits of economic agents, thereby offering relevant derived policy suggestions. The first part of this thesis is dedicated to the analysis of financial markets and traders who populate them. Firstly, psychological and behavioural characteristics are examined, with a focus on traders’ experience. By means of an analogy between human psychomotor skills in balancing unstable objects and their ability in resisting destabilising financial market signals, this study demonstrates that the personal experiences of market participants impact the system’s stability. Moreover, the similitude between the human nervous system attempting to balance a stick on a fingertip (i.e., human stick-balancing) and financial markets is further explored by considering the latter as a single entity which tries to maintain asset prices in equilibrium. In this framework, we demonstrate that, by introducing a small amount of noise, intermittent changes in the stick’s angular velocity are able to replicate the most famous stylized facts of financial returns. Subsequently, the analysis proceeds by focusing on traders’ abilities to make investment decisions. Specifically, we compare different portfolio optimization techniques, such as maximizing the Sharpe Ratio, minimizing the Expected Shortfall, and employing "zero-intelligence" trading. Empirical financial time series and simulated series generated by an order-book model are used to assess how these strategies affect both individual performances and market dynamics. Finally, changing context, we investigate how heterogeneous features of the labour force may influence macroeconomic aggregate results. Workers are distinguished based on their skills and productivity, which are influenced by their educational background. We model two possible education courses: one more general and one "vocational", in which students acquire specific skills directly required by firms. Findings underscore the significant impact the general educated workers should have on a country’s macroeconomic performance.
Nei sistemi complessi, la dinamica aggregata è generata dall'interazione tra i singoli costituenti e presenta proprietà che non possono essere dedotte osservando gli attributi di ciascuno di essi. Tuttavia, poiché le caratteristiche di questi individui inducono le loro decisioni, influenzando così le conseguenti interazioni, anche se è impossibile stabilire la linea di causalità tra le singole scelte e le proprietà aggregate del sistema, c'è spazio per un'indagine sulla variazione indotta sulle dinamiche emergenti da cambiamenti nella caratterizzazione degli individui. Pertanto, per mezzo di modelli ad agenti, è possibile studiare come i fenomeni emergenti delle dinamiche macroeconomiche cambino in risposta alle modifiche delle distribuzioni statistiche dei singoli tratti degli agenti economici, fornendo così rilevanti suggerimenti di possibili politiche economiche. La prima parte di questa tesi è dedicata all'analisi dei mercati finanziari e dei traders che li popolano. In primo luogo, vengono esaminate le caratteristiche psicologiche e comportamentali, con particolare attenzione all'esperienza dei traders. Attraverso un'analogia tra le abilità psicomotorie umane nel bilanciare oggetti instabili e la loro capacità di resistere ai segnali destabilizzanti dei mercati finanziari, questo studio dimostra che le esperienze personali dei partecipanti al mercato hanno un impatto sulla stabilità del sistema. Inoltre, la similitudine tra il sistema nervoso umano che tenta di bilanciare una stecca su un polpastrello (cioè lo “human stick-balancing”) e i mercati finanziari viene ulteriormente esplorata considerando questi ultimi come un'unica entità che cerca di mantenere i prezzi degli assets in equilibrio. In questo contesto, dimostriamo che, introducendo una piccola quantità di rumore, i cambiamenti intermittenti nella velocità angolare della stecca sono in grado di replicare i più famosi fatti stilizzati dei rendimenti finanziari. Successivamente, l'analisi procede concentrandosi sulle capacità dei traders di prendere decisioni di investimento. In particolare, vengono confrontate diverse tecniche di ottimizzazione del portafoglio, come la massimizzazione dello Sharpe Ratio, la minimizzazione dell'Expected Shortfall e l'impiego di un trading ad "intelligenza zero", cioè agire in modo casuale. Le serie finanziarie empiriche e le serie simulate generate da un modello di order-book sono utilizzate per valutare come queste strategie influenzino sia le performance individuali sia le dinamiche di mercato. Infine, cambiando contesto, investighiamo su come le caratteristiche eterogenee della forza lavoro possano influenzare i risultati macroeconomici aggregati. I lavoratori sono distinti in base alle loro competenze e alla loro produttività, che sono influenzate dal loro percorso formativo. Modelliamo due possibili percorsi di istruzione: uno più generale e uno "vocational" (professionale), in cui gli studenti acquisiscono competenze specifiche direttamente richieste dalle imprese. I risultati sottolineano l'impatto significativo che i lavoratori con un'istruzione generale dovrebbero avere sulla performance macroeconomica di un paese.
Un'indagine sul ruolo della Complessità nei sistemi Macroeconomici e Finanziari
MAZZARINO, LAURA
2024
Abstract
In complex systems, the aggregate dynamics is generated by the interaction among their individual constituents and it exhibits properties that cannot be inferred by observing attributes of each of them. However, since the features of those individuals induce their decisions, thus affecting their consequent interactions, although it is impossible to establish the causation line between single choices and the aggregate properties of the system, there is room for an investigation on the variation induced on the emergent dynamics by changes in the characterization of individuals. Thus, by means of agent-based models, it is possible to investigate how emergent phenomena of macroeconomic dynamics change in response to modification in the statistical distributions of individual traits of economic agents, thereby offering relevant derived policy suggestions. The first part of this thesis is dedicated to the analysis of financial markets and traders who populate them. Firstly, psychological and behavioural characteristics are examined, with a focus on traders’ experience. By means of an analogy between human psychomotor skills in balancing unstable objects and their ability in resisting destabilising financial market signals, this study demonstrates that the personal experiences of market participants impact the system’s stability. Moreover, the similitude between the human nervous system attempting to balance a stick on a fingertip (i.e., human stick-balancing) and financial markets is further explored by considering the latter as a single entity which tries to maintain asset prices in equilibrium. In this framework, we demonstrate that, by introducing a small amount of noise, intermittent changes in the stick’s angular velocity are able to replicate the most famous stylized facts of financial returns. Subsequently, the analysis proceeds by focusing on traders’ abilities to make investment decisions. Specifically, we compare different portfolio optimization techniques, such as maximizing the Sharpe Ratio, minimizing the Expected Shortfall, and employing "zero-intelligence" trading. Empirical financial time series and simulated series generated by an order-book model are used to assess how these strategies affect both individual performances and market dynamics. Finally, changing context, we investigate how heterogeneous features of the labour force may influence macroeconomic aggregate results. Workers are distinguished based on their skills and productivity, which are influenced by their educational background. We model two possible education courses: one more general and one "vocational", in which students acquire specific skills directly required by firms. Findings underscore the significant impact the general educated workers should have on a country’s macroeconomic performance.File | Dimensione | Formato | |
---|---|---|---|
Mazzarino-PhD-Tesi.pdf
embargo fino al 07/06/2025
Dimensione
43.43 MB
Formato
Adobe PDF
|
43.43 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/165728
URN:NBN:IT:UNICT-165728