Análisis de las estrategias de producción y del precio en un mercado competitivo
Loading...
Download
Official URL
Full text at PDC
Publication date
2025
Defense date
02/2025
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citation
Abstract
Este Trabajo de Fin de Máster analiza el mercado de los diamantes desde el punto de vista de la teoría de juegos y de las técnicas de machine learning. Por un lado, se trata de ver si las principales empresas del mercado cooperan entre sí o actúan por su cuenta. Todas actúan según el equilibrio de Nash y no hay motivos para pensar en una posible colusión.
Por otro lado, se predice el precio del diamante en función de la cantidad agregada al mercado y la fecha. Estas son las variables que tiene la base de datos utilizada junto a las que se pueden obtener de ellas como la cantidad agregada el día o año anterior. Se utilizan para ello algoritmos como Regresión Lineal, Árbol de Decisión, Bagging y Random Forest. La medida de bondad de ajuste utilizada para su comparación es el RMSE y el mejor algoritmo para los datos señalados es Regresión Lineal.
This TFM analyzes the diamond market from the perspective of game theory and machine learning techniques. On one hand, it examines whether the main companies in the market cooperate with each other or act independently. All of them act according to the Nash equilibrium, and there is no reason to suspect any potential collusion. On the other hand, the price of diamonds is predicted based on the amount added to the market and the date. These are the variables in the database used, along with those that can be derived from them, such as the amount added the previous day or year. Algorithms like Linear Regression, Decision Trees, Bagging, and Random Forest are used for this purpose. The goodness-of-fit measure used for comparison is RMSE, and the best algorithm for the specified data is Linear Regression
This TFM analyzes the diamond market from the perspective of game theory and machine learning techniques. On one hand, it examines whether the main companies in the market cooperate with each other or act independently. All of them act according to the Nash equilibrium, and there is no reason to suspect any potential collusion. On the other hand, the price of diamonds is predicted based on the amount added to the market and the date. These are the variables in the database used, along with those that can be derived from them, such as the amount added the previous day or year. Algorithms like Linear Regression, Decision Trees, Bagging, and Random Forest are used for this purpose. The goodness-of-fit measure used for comparison is RMSE, and the best algorithm for the specified data is Linear Regression