Bandwidth Selection for Prediction in Regression

  1. Inés Barbeito 1
  2. Ricardo Cao 1
  3. Stefan Sperlich 2
  1. 1 Universidade da Coruña
    info

    Universidade da Coruña

    La Coruña, España

    ROR https://ror.org/01qckj285

  2. 2 Université de Genève
    info

    Université de Genève

    Ginebra, Suiza

    ROR https://ror.org/01swzsf04

Libro:
XoveTIC 2019: The 2nd XoveTIC Conference (XoveTIC 2019), A Coruña, Spain, 5–6 September
  1. Alberto Alvarellos González (ed. lit.)
  2. José Joaquim de Moura Ramos (ed. lit.)
  3. Beatriz Botana Barreiro (ed. lit.)
  4. Javier Pereira Loureiro (ed. lit.)
  5. Manuel F. González Penedo (ed. lit.)

Editorial: MDPI

ISBN: 978-3-03921-444-0,  978-3-03921-443-3

Año de publicación: 2019

Congreso: XoveTIC (2. 2019. A Coruña)

Tipo: Aportación congreso

DIALNET GOOGLE SCHOLAR lock_openAcceso abierto editor

Resumen

There exist many different methods to choose the bandwidth in kernel regression. If, however, the target is regression based prediction for samples or populations with potentially different distributions, then the existing methods can easily be suboptimal. This situation occurs for example in impact evaluation, data matching, or scenario simulations. We propose a bootstrap method to select a global bandwidth for nonparametric out-of-sample prediction. The asymptotic theory is developed, and simulation studies show the successful operation of our method. The method is used to predict nonparametrically the salary of Spanish women if they were paid along the same wage equation as men, given their own characteristics.

Fuente de los datos: Dialnet