The best choice for predicting Intel Corporation's peak stock price: Regression Tree or Multiple Linear Regression?

• Autores: Carpio-Vargas, Edgar Eloy, Mayda-Huanca, Alicia Roxana, Espinoza-Rivas, German Rafael, Mamani Vilca, Ecler, Ccolqque Ruiz, Betsabe Milagros
• Formato: artículo
• Fecha de Publicación: 2023
• Institución: Universidad Nacional Micaela Bastidas de Apurímac
• Repositorio: UNMB-Riqchary
• Lenguaje: español
• OAI Identifier: oai:revistas.unamba.edu.pe:article/25
• Enlace del recurso: https://revistas.unamba.edu.pe/index.php/riqchary/article/view/25
• Nivel de acceso: acceso abierto
• Materia: Árboles de regresión; regresión lineal multiple; Multiple Linear Regression; Regression Trees

The objective of the study was to compare the performance of the Regression Tree versus the Multiple Linear Regression Model in relation to the opening price and daily sales volume of Intel Corporation shares. A descriptive correlational non-experimental correlational research with cross-sectional design was conducted using a convenience sample. The sample consisted of 410 records collected from May 2018 to October 2019, obtained through documentary review. The results obtained showed that the Regression Tree established that the most significant variable to explain the maximum stock price was the opening price, discarding the volume variable. The Mean Squared Error obtained was $1.4480. On the other hand, the multiple linear regression model, using the outlier elimination technique, presented a Residual Standard Error of 0.2257 dollars. In conclusion, it was determined that the most adequate model to predict the maximum price of Intel Corporation shares is the Multiple Linear Regression Model with the elimination of outlier points.