The thesis has the purpose of validating the use of Python Prophet as a tool to perform oil price forecasting. The research is based on the prevision of three months for oil price by using historical data retrieved from Bloomberg official website. The oil industry is known to be complex and characterized by erratic trends, two features that make the analysis challenging and, through the research, demonstrate the user friendliness of the tool. The project is really relevant because companies too often face unexpected expenses due to the price of commodity and, in the case of oil, the related transportation costs, in every business in which theres logistic or shipment, may erode margins and ruin financial controlling and budgeting prevision. We have the ultimate goal of proving that with an appropriate tool, effective forecasting can be successfully performed and can help the analyst in taking informed and precise business decisions. The whole research is run following the Design Science Research methodology identifying the main problems, defining relevant metarequirements and design principles and providing a final solution.
La tesi ha lobiettivo di validare luso di Python Prophet come strumento per attuare previsioni del prezzo del petrolio. La ricerca è basata sulla previsione del prezzo del petrolio per un periodo di tre mesi utilizzando dati storici del sito ufficiale di Bloomberg. Lindustria petrolifera è conosciuta per essere complessa e caratterizzata da andamenti irregolari, due caratteristiche che rendono la nostra analisi sfidante, e attraverso la ricerca, dimostrano la facilità duso per lutente del nostro strumento. Il progetto è veramente rilevante perché le aziende troppo spesso affrontano spese inaspettate dovute al prezzo delle merci e, nel caso del petrolio, gli annessi costi di trasporto in ogni business nel quale vi è logistica o spedizioni può erodere margini e rovinare previsioni di controllo finanziario e di budget. Noi abbiamo lobiettivo di provare che con lo strumento appropriato, previsioni efficaci possono essere attuate che possono aiutare lanalista a prendere informate e precise decisioni per il business. Lintera ricerca è stata redatta seguendo la metodologia del Design Science Research, definendo rilevanti metarequirements e design principles e fornendo una soluzione finale.
VALIDATING THE USE OF PYTHON PROPHET AS A TOOL FOR OIL FORECASTING: OPPORTUNITIES AND RISKS FOR BUSINESSES
COGHETTO, MATTIA
2018/2019
Abstract
The thesis has the purpose of validating the use of Python Prophet as a tool to perform oil price forecasting. The research is based on the prevision of three months for oil price by using historical data retrieved from Bloomberg official website. The oil industry is known to be complex and characterized by erratic trends, two features that make the analysis challenging and, through the research, demonstrate the user friendliness of the tool. The project is really relevant because companies too often face unexpected expenses due to the price of commodity and, in the case of oil, the related transportation costs, in every business in which theres logistic or shipment, may erode margins and ruin financial controlling and budgeting prevision. We have the ultimate goal of proving that with an appropriate tool, effective forecasting can be successfully performed and can help the analyst in taking informed and precise business decisions. The whole research is run following the Design Science Research methodology identifying the main problems, defining relevant metarequirements and design principles and providing a final solution.È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per maggiori informazioni e per verifiche sull'eventuale disponibilità del file scrivere a: unitesi@unipv.it.
https://hdl.handle.net/20.500.14239/5734