This thesis proposes a methodological framework for the valuation of a company’s equity security, applying the Discounted Cash Flow (DCF) model and focusing on the crucial issue of forecasting future Free Cash Flow Per Share (FCFPS). To this end, econometric and machine learning models are implemented to leverage a set of explanatory variables with predictive power, in order to generate, over an extended time horizon, a sequence of expected cash flows to be used in the DCF formula. The adopted approach allows overcoming the limitations of traditional 5-year projections, enabling more flexible and informative estimates. The resulting forecasts are then incorporated into a Monte Carlo simulation framework, with the aim of deriving a probabilistic distribution of the stock’s fair value. The model is further extended with a Bayesian component, allowing the integration of the analyst’s subjective opinion in the form of probability distributions. The framework is applied to the case study of International Business Machines Corporation (IBM), ultimately producing a distribution of the stock’s fair value, which is useful for supporting decisions under uncertainty.
La presente tesi propone un framework metodologico per la valutazione del titolo azionario rappresentativo del capitale di rischio di un’impresa, applicando il modello Discounted Cash Flow (DCF) e concentrandosi sul nodo cruciale della previsione dei Free Cash Flow Per Share (FCFPS) futuri. A tal fine, vengono implementati modelli econometrici e di machine learning capaci di sfruttare un insieme di variabili esplicative con potere predittivo, al fine di generare, su un orizzonte temporale esteso, una sequenza di flussi di cassa attesi da impiegare nella formula del DCF. L’approccio adottato consente di superare i limiti imposti dalle tradizionali proiezioni a 5 anni, permettendo stime più flessibili e informative. Le previsioni ottenute vengono poi inserite in un framework simulativo Monte Carlo, con l’obiettivo di derivare una distribuzione probabilistica del fair value del titolo. Il modello è infine esteso con una componente bayesiana, utile a integrare l’opinione soggettiva dell’analista sotto forma di distribuzioni di probabilità. Il framework viene applicato al caso studio di International Business Machines Corporation (IBM), restituendo come output una distribuzione del valore equo del titolo, utile per supportare decisioni in contesti di incertezza.
Valutazione probabilistica del Fair Value Azionario: un’applicazione estesa del modello DCF a IBM
PIUNTI, LORENZO
2024/2025
Abstract
This thesis proposes a methodological framework for the valuation of a company’s equity security, applying the Discounted Cash Flow (DCF) model and focusing on the crucial issue of forecasting future Free Cash Flow Per Share (FCFPS). To this end, econometric and machine learning models are implemented to leverage a set of explanatory variables with predictive power, in order to generate, over an extended time horizon, a sequence of expected cash flows to be used in the DCF formula. The adopted approach allows overcoming the limitations of traditional 5-year projections, enabling more flexible and informative estimates. The resulting forecasts are then incorporated into a Monte Carlo simulation framework, with the aim of deriving a probabilistic distribution of the stock’s fair value. The model is further extended with a Bayesian component, allowing the integration of the analyst’s subjective opinion in the form of probability distributions. The framework is applied to the case study of International Business Machines Corporation (IBM), ultimately producing a distribution of the stock’s fair value, which is useful for supporting decisions under uncertainty.File | Dimensione | Formato | |
---|---|---|---|
Tesi IBM.pdf
accesso aperto
Dimensione
1.89 MB
Formato
Adobe PDF
|
1.89 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.12075/21976