Regresión no paramétrica
Loading...
Identifiers
Publication date
Authors
Advisors
Tutors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
La regresión se ocupa de analizar cómo influye una variable X, llamada independiente, sobre otra Y , llamada dependiente. Hay dos enfoques distintos: la regresión paramétrica y la no
paramétrica.
En este trabajo se estudian y comparan ambas, al estimar el modelo de regresión el enfoque
no paremátrico es una alternativa a los modelos clásicos de regresión lineal ya que aportan mucha
flexibilidad en la forma del modelo.
Ya desde la perspectiva de la regresión no paramétrica, el estudio se centra en el cálculo
del estimador de Nadaraya-Watson. Este estimador dependerá de un parámetro, denominado
ventana. Estudiaremos la influencia que tiene el parámetro ventana con el objetivo de encontrar
la ventana que minimice el error cuadrático medio del estimador para así conseguir el más
eficiente.
Por último, se ilustra la aplicación práctica de los métodos presentados, abordando la estimación no paramétrica de la función de regresión de un conjunto de datos reales y en particular, el
cálculo de la ventana óptima del estimador. Dichos datos pueden llegar a ser de interés para que
los equipos de baloncesto puedan mejorar la siguiente temporada incorporando en su plantilla a
los jugadores más valorados de la liga.
This regression is concerned with analyzing how a variable X, called independent, influences another variable Y , called dependent. There are two different approaches: parametric and nonparametric regression. In this paper we study and compare the two. In estimating the regression model, the nonparametric approach is an alternative to the classical linear regression models since it provides much flexibility in the form of the model. From the nonparametric regression perspective, the study focuses on the calculation of the Nadaraya-Watson estimator. This estimator will depend on a parameter, called bandwidth. We will study the influence of the bandwidth parameter with the aim of finding the bandwidth that minimizes the mean square error of the estimator in order to obtain the most eficient one. Finally, the practical application of the methods presented is illustrated, addressing the nonparametric estimation of the regression function of a set of real data and in particular, the calculation of the optimal bandwidth of the estimator. Such data may be of interest for basketball teams to improve the following season by incorporating in their roster the most valued players in the league.
This regression is concerned with analyzing how a variable X, called independent, influences another variable Y , called dependent. There are two different approaches: parametric and nonparametric regression. In this paper we study and compare the two. In estimating the regression model, the nonparametric approach is an alternative to the classical linear regression models since it provides much flexibility in the form of the model. From the nonparametric regression perspective, the study focuses on the calculation of the Nadaraya-Watson estimator. This estimator will depend on a parameter, called bandwidth. We will study the influence of the bandwidth parameter with the aim of finding the bandwidth that minimizes the mean square error of the estimator in order to obtain the most eficient one. Finally, the practical application of the methods presented is illustrated, addressing the nonparametric estimation of the regression function of a set of real data and in particular, the calculation of the optimal bandwidth of the estimator. Such data may be of interest for basketball teams to improve the following season by incorporating in their roster the most valued players in the league.
Description
Traballo Fin de Grao en Matemáticas. Curso 2021-2022
Keywords
Bibliographic citation
Relation
Has part
Has version
Is based on
Is part of
Is referenced by
Is version of
Requires
Sponsors
Rights
Atribución-NoComercial-CompartirIgual 4.0 Internacional



