Publication: Comparación entre XGBoost y Regresión Lineal Múltiple para la predicción de la evolución del precio de las acciones
Loading...
Official URL
Full text at PDC
Publication Date
2022-06
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
La predicción de la evolución del precio de las acciones de la bolsa de valores es un estudio que lleva practicándose durante mucho tiempo. El desarrollo del área del Data Science y de los algoritmos de Machine Learning abre nuevas posibilidades para llevar a cabo predicciones como esta. Este trabajo se centra en el desarrollo de un proyecto de Data Science siguiendo una metodología adecuada para el mismo. Con este proyecto se elaboran dos modelos predictivos para los precios de las acciones: uno con XGBoost y otro con Regresión Lineal Múltiple. De esta manera, se pretende comparar la eficacia de cada uno de los algoritmos para el caso que se estudia y poder determinar, por lo tanto, cuál de ellos resulta más útil para la predicción del precio de las acciones.
Stock price evolution prediction is a study that has been practiced for a long time. The development of the area of Data Science and Machine Learning algorithms opens up new possibilities to carry out predictions like this. This piece of work focuses on the development of a Data Science project following an appropriate methodology for it. With this project, two predictive models for stock prices are developed: one with XGBoost and the other with Multiple Linear Regression. In this way, it is intended to compare the effectiveness of each of the algorithms for the case under study and to be able to determine, therefore, which of them is more useful for stock price prediction.
Stock price evolution prediction is a study that has been practiced for a long time. The development of the area of Data Science and Machine Learning algorithms opens up new possibilities to carry out predictions like this. This piece of work focuses on the development of a Data Science project following an appropriate methodology for it. With this project, two predictive models for stock prices are developed: one with XGBoost and the other with Multiple Linear Regression. In this way, it is intended to compare the effectiveness of each of the algorithms for the case under study and to be able to determine, therefore, which of them is more useful for stock price prediction.
Description
Trabajo de Fin de Grado en Ingeniería Informática, Facultad de Informática UCM, Departamento de Sistemas Informáticos y Computación, Curso 2021/2022.