El método de alzado sucesivo y su relación con el método cresta

  1. García Pérez, José
  2. García García, Catalina Beatriz
  3. López Martín, María del Mar
  4. Salmerón Gómez, Román
Libro:
Anales de economía aplicada 2014
  1. García Lizana, Antonio (coord.)
  2. Fernández Morales, Antonio (coord.)
  3. Podadera Rivera, Pablo (coord.)

Editorial: Asociación Española de Economía Aplicada, ASEPELT

Ano de publicación: 2014

Páxinas: 1323-1338

Congreso: ASEPELT España. Reunión anual (28. 2014. Málaga)

Tipo: Achega congreso

Resumo

The collinearity problem deals with the existence of linear relationships between two or more independent variables in a linear model. Some of the possible solutions can be to improve the sample or simply to dispense with the variable that produces collinearity. The ridge regression was proposed to treat this problem in a more mechanical and objective way. Alternatively, García et al 2010 have proposed the raise regression to estimate a linear model with collinearity. In this paper we show that the ridge estimator can be obtained from the successive application of the raise regression under some conditions. We illustrate our results with an empirical application.