Título : Análisis de la relación dinámica entre la comunicación en X y los CDs |
Autor : García Martínez, Miguel Ángel |
Tutor: Gimeno Blanes, Francisco Javier  Rodríguez Ibáñez, Margarita  |
Editor : Universidad Miguel Hernández de Elche |
Departamento: Departamentos de la UMH::Ingeniería de Comunicaciones |
Fecha de publicación: 2024-09 |
URI : https://hdl.handle.net/11000/33914 |
Resumen :
En la era de la información, el análisis de datos y la predicción de comportamientos en los
mercados financieros se han vuelto herramientas esenciales para inversores y analistas. Este
proyecto se enfoca en desarrollar un modelo predictivo que analice la relación entre la
actividad en X (Twitter) y... Ver más
In the information age, data analysis and the prediction of behaviour in financial markets have
become essential tools for investors and analysts. This project focuses on developing a
predictive model that analyses the relationship between activity on X (Twitter) and the values
of Credit Default Swaps (CDs) of various financial institutions.
This study focuses on four European banks: Credit Suisse, Deutsche Bank, Commerzbank and
Banca Monte dei Paschi di Siena, institutions that have faced significant financial challenges in
recent years. The analysis uses advanced Natural Language Processing (NLP) and machine
learning (ML) techniques in Matlab, exploring how social media activity can impact the
perception of financial risk, as reflected in Credit Default Swaps (CDS).
To conduct this research, we used historical CD data extracted from the DataStream database
with tweets obtained through the former Twitter API and the Graphext platform. The study
period spans from January 2017 to May 2023.
The development of the predictive model follows several key steps. First, a sentiment analysis
of the tweets was performed, classifying them as positive, negative or neutral, which allows us
to quantify the emotions expressed on social networks with respect to the companies
analysed. Subsequently, this data was integrated with additional variables such as user account
verification and number of followers to improve the relevance of the analysis. Finally, historical
information from the CDS was used to train the model and improve the accuracy of the
predictions.
The use of Matlab as the central tool of the project facilitates the processing of large volumes
of data and the execution of complex prediction algorithms, given its capacity to handle
intensive mathematical calculations and its wide range of analysis tools. The methodological
approach applied includes tokenisation and normalisation of tweets, frequency and intensity
analysis, and the implementation of predictive models based on statistical and machine
learning techniques.
This project aims to provide a solid basis for future research in the area of the correlation
between social networks and Credit Default Swaps, exploring how communication on X can
have an impact on the perception of credit risk of financial institutions. In addition, the study
offers a useful tool for understanding the relationship between critical events and CDS
fluctuations, in order to predict abnormal behaviour in financial markets.
|
Palabras clave/Materias: redes sociales X (Twitter) Creadit Default Swaps (CDS) Procesamiento del Lenguaje Natural (NLP) aprendizaje automático (machine learning) |
Área de conocimiento : CDU: Ciencias aplicadas: Ingeniería. Tecnología |
Tipo de documento : info:eu-repo/semantics/bachelorThesis |
Derechos de acceso: info:eu-repo/semantics/openAccess Attribution-NonCommercial-NoDerivatives 4.0 Internacional |
Aparece en las colecciones: TFG- Ingeniería de Tecnologías de Telecomunicación
|