My doctoral thesis merges two main streams of research. The first one is tied to the accounting field and uses the Covid-19 pandemic as research setting. For instance, the incumbency of the Covid-19 pandemic at the early stages of 2020 starts a phase of heightened uncertainty in global financial markets, such that the usual state of thing does no longer hold. Most importantly, it is a shared belief among academics and practitioners that the pandemic has been unlike anything experienced in the past, involving unprecedented and unexperienced consequences, at least from an economic point of view. Having this considered, an effort by researchers is deemed necessary to understand the new mechanisms regulating a pandemic-affected market and to rethink the well consolidated knowledge under the new lights of this turbulent market phase. In my doctoral thesis, we contribute to the Covid19-related accounting literature with two research papers (which are Chapter 1 and 2 of the thesis respectively). These two papers investigate empirically the extent to which the accounting informs the market about the underlying firms’ fundamental during phases characterized by heightened fundamental uncertainty (derived from the Covid-19 pandemic). The second stream of research is more tied to the corporate finance field. Over the recent years, the spread of machine learning based techniques has enabled researchers to reframe many economic or financial problems exploiting the advantages brought by these new techniques, reaching far better solutions than traditional models. In Chapter 3 of my doctoral thesis I introduce the use of machine learning into the accounting-based default prediction literature, advancing also the financial knowledge processed inside the models following a data-analytic approach.
La mia tesi di dottorato combina principalmente due filoni di ricerca. Il primo è strettamente legato al campo dell’economia aziendale e utilizza la pandemia di Covid-19 come contesto sperimentale. Infatti, l’avvento della pandemia ad inizio 2020 ha iniziato una fase di forte incertezza nei mercati finanziari mondiali, a livelli tali da stravolgerne il loro normale funzionamento. Soprattutto, è opinione diffusa da accademici e addetti ai lavori che la recente pandemia di Covid-19 sia stata estremamente diversa da qualsiasi altro evento simile accaduto in passato, ed abbia comportato conseguenze imprevedibili e mai sperimentate fino ad ora (almeno da un punto di vista economico-finanziario). Avendo considerato ciò, un particolare sforzo è stato richiesto al mondo della ricerca per ripensare a tutta la conoscenza consolidata in ambito economico sotto i nuovi riflettori della pandemia. Nella mia tesi di dottorato, contribuisco alla letteratura collegata alla pandemia di Covid-19 con due lavori (che costituiscono il Capitolo 1 e 2 della tesi rispettivamente). Queste due produzioni indagano empiricamente come l’informazione contabile e finanziaria abbia spiegato il sottostante economico delle aziende durante una fase di mercato caratterizzata da incertezza estrema, derivata dall’avvento della pandemia. Il secondo filone di ricerca a cui riferisce questa tesi è legato alla finanza aziendale. Recentemente, la diffusione delle tecniche di predizione e classificazione di tipo Machine Learning ha consentito ai ricercatori di riformulare molti problemi storici di tipo economico o finanziario sfruttando i vantaggi portati da queste nuove tecniche, raggiungendo risultati migliori e più soddisfacenti rispetto ai modelli tradizionali. Nel Capitolo 3 della tesi, introduco l’utilizzo delle tecniche di tipo Machine Learning nel problema economico-finanziario della predizione di un futuro stato di bancarotta. Allo stesso tempo, provo ad evolvere la teoria processata dai modelli stessi seguendo un approccio data-analitico.
IL RUOLO DELL'INFORMATIVA CONTABILE DURANTE CIRCOSTANZE STRAORDINARIE.
LONGHIN, FEDERICO
2023
Abstract
My doctoral thesis merges two main streams of research. The first one is tied to the accounting field and uses the Covid-19 pandemic as research setting. For instance, the incumbency of the Covid-19 pandemic at the early stages of 2020 starts a phase of heightened uncertainty in global financial markets, such that the usual state of thing does no longer hold. Most importantly, it is a shared belief among academics and practitioners that the pandemic has been unlike anything experienced in the past, involving unprecedented and unexperienced consequences, at least from an economic point of view. Having this considered, an effort by researchers is deemed necessary to understand the new mechanisms regulating a pandemic-affected market and to rethink the well consolidated knowledge under the new lights of this turbulent market phase. In my doctoral thesis, we contribute to the Covid19-related accounting literature with two research papers (which are Chapter 1 and 2 of the thesis respectively). These two papers investigate empirically the extent to which the accounting informs the market about the underlying firms’ fundamental during phases characterized by heightened fundamental uncertainty (derived from the Covid-19 pandemic). The second stream of research is more tied to the corporate finance field. Over the recent years, the spread of machine learning based techniques has enabled researchers to reframe many economic or financial problems exploiting the advantages brought by these new techniques, reaching far better solutions than traditional models. In Chapter 3 of my doctoral thesis I introduce the use of machine learning into the accounting-based default prediction literature, advancing also the financial knowledge processed inside the models following a data-analytic approach.File | Dimensione | Formato | |
---|---|---|---|
Thesis Federico Longhin.pdf
embargo fino al 10/04/2026
Dimensione
1.37 MB
Formato
Adobe PDF
|
1.37 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/80543
URN:NBN:IT:UNIPD-80543