Modelos de regresión con penalizacións
Loading...
Identifiers
Publication date
Authors
Advisors
Tutors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
[ES] Los modelos de regresión se basan en los estimadores de mínimos cuadrados, el cual
veremos que es fácil que no sea capaz de estimar los coeficientes con fiabilidad en el caso
de variables dependientes. Para corregir este problema, se crearon distintos métodos que
veremos en este trabajo. La Regresión Ridge y el método LASSO utilizan el estimador de
mínimos cuadrados con una penalización, la cual restringe los posibles valores que pueden
tomar los coeficientes, y así logran controlar sus valores. Por otro lado, LAR nos da una
ordenación de las variables según su importancia, con la cual podemos crear modelos de
regresión cogiendo sólo variables importantes. Con estos métodos conseguimos evitar el
problema de las variables dependientes.
[EN] Regression methods are based on the least square estimator, which may not provide a good estimation of the coefficients in the case we have dependant variables. To avoid this problem, different methods were created. Ridge Regression and LASSO use the least square estimator with a penalty that shrinks the values of the coefficients, controling their possible values. On the other hand, LAR order the variables by their importance in the model, and we can use it to create a model by choosing only the most important variables. With these methods we can avoid the problem we had in the case of dependant variables.
[EN] Regression methods are based on the least square estimator, which may not provide a good estimation of the coefficients in the case we have dependant variables. To avoid this problem, different methods were created. Ridge Regression and LASSO use the least square estimator with a penalty that shrinks the values of the coefficients, controling their possible values. On the other hand, LAR order the variables by their importance in the model, and we can use it to create a model by choosing only the most important variables. With these methods we can avoid the problem we had in the case of dependant variables.
Description
Traballo Fin de Grao en Matemáticas. Curso 2018-2019
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



