Additive nonparametric regression estimation via back tting and marginal integration under common bandwidth selection criterion : small sample performance
In this paper, we conducted a Monte Carlo investigation to reveal some charac- teristics of nite sample distributions of the back tting (B) and Marginal Integration (MI) estimators for an additive bivariate regression. We are particularly interested in providing some evidence on how the di¤erent me...
Main Author: | |
---|---|
Other Authors: | |
Format: | Others |
Language: | Portuguese |
Published: |
2015
|
Subjects: | |
Online Access: | http://hdl.handle.net/10183/109669 |
id |
ndltd-IBICT-oai-lume.ufrgs.br-10183-109669 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-IBICT-oai-lume.ufrgs.br-10183-1096692018-10-21T17:04:35Z Additive nonparametric regression estimation via back
tting and marginal integration under common bandwidth selection criterion : small sample performance Silva, Fernando Augusto Boeira Sabino da Sen, Pranab Kumar Economia Métodos qualitativos Modelos matemáticos Econometria In this paper, we conducted a Monte Carlo investigation to reveal some charac- teristics of
nite sample distributions of the back
tting (B) and Marginal Integration (MI) estimators for an additive bivariate regression. We are particularly interested in providing some evidence on how the di¤erent methods for the selection of bandwidth, such as the plug-in method, inuence the
nite sample properties of the MI and B estimators. We are particularly concerned with the performance of these estimators when bandwidth selection is done based in data driven methods, since in this case the aymptotics properties of these estimators are currently unavailable. The impact of ignoring the dependency between regressors is also investigated. Finally, di¤erently from what occurs at the present time, when the B and MI estimators are used ad-hoc, our objective is to provide information that allows for a more accurate comparison of these two competing alternatives in a
nite sample setting. 2015-02-05T02:17:13Z 2006 info:eu-repo/semantics/publishedVersion info:eu-repo/semantics/masterThesis http://hdl.handle.net/10183/109669 000951323 por info:eu-repo/semantics/openAccess application/pdf reponame:Biblioteca Digital de Teses e Dissertações da UFRGS instname:Universidade Federal do Rio Grande do Sul instacron:UFRGS |
collection |
NDLTD |
language |
Portuguese |
format |
Others
|
sources |
NDLTD |
topic |
Economia Métodos qualitativos Modelos matemáticos Econometria |
spellingShingle |
Economia Métodos qualitativos Modelos matemáticos Econometria Silva, Fernando Augusto Boeira Sabino da Additive nonparametric regression estimation via back tting and marginal integration under common bandwidth selection criterion : small sample performance |
description |
In this paper, we conducted a Monte Carlo investigation to reveal some charac- teristics of
nite sample distributions of the back
tting (B) and Marginal Integration (MI) estimators for an additive bivariate regression. We are particularly interested in providing some evidence on how the di¤erent methods for the selection of bandwidth, such as the plug-in method, inuence the
nite sample properties of the MI and B estimators. We are particularly concerned with the performance of these estimators when bandwidth selection is done based in data driven methods, since in this case the aymptotics properties of these estimators are currently unavailable. The impact of ignoring the dependency between regressors is also investigated. Finally, di¤erently from what occurs at the present time, when the B and MI estimators are used ad-hoc, our objective is to provide information that allows for a more accurate comparison of these two competing alternatives in a
nite sample setting. |
author2 |
Sen, Pranab Kumar |
author_facet |
Sen, Pranab Kumar Silva, Fernando Augusto Boeira Sabino da |
author |
Silva, Fernando Augusto Boeira Sabino da |
author_sort |
Silva, Fernando Augusto Boeira Sabino da |
title |
Additive nonparametric regression estimation via back
tting and marginal integration under common bandwidth selection criterion : small sample performance |
title_short |
Additive nonparametric regression estimation via back
tting and marginal integration under common bandwidth selection criterion : small sample performance |
title_full |
Additive nonparametric regression estimation via back
tting and marginal integration under common bandwidth selection criterion : small sample performance |
title_fullStr |
Additive nonparametric regression estimation via back
tting and marginal integration under common bandwidth selection criterion : small sample performance |
title_full_unstemmed |
Additive nonparametric regression estimation via back
tting and marginal integration under common bandwidth selection criterion : small sample performance |
title_sort |
additive nonparametric regression estimation via back
tting and marginal integration under common bandwidth selection criterion : small sample performance |
publishDate |
2015 |
url |
http://hdl.handle.net/10183/109669 |
work_keys_str_mv |
AT silvafernandoaugustoboeirasabinoda additivenonparametricregressionestimationviabackttingandmarginalintegrationundercommonbandwidthselectioncriterionsmallsampleperformance |
_version_ |
1718782790984531968 |