Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

XTBCFE: Stata module to perform bootstrap-corrected Fixed Effects estimation and inference in dynamic panel models

De Vos, Ignace LU ; Everaert, Gerdie and Ruyssen, Ilse (2015)
Abstract
xtbcfe is a bootstrap-corrected fixed effects (LSDV) estimator for dynamic panel data models of general order. It estimates the specified model with the fixed effects estimator and corrects its small T bias (see Nickell, 1981) using a simplified but extended version of the approach presented in Everaert and Pozzi (2007). The algorithm evaluates the bias of the fixed effects estimator in a numerical way to avoid the use of analytical correction formulas. xtbcfe is therefore also applicable to higher order (more than one lag of the dependent variable) models with a potential non-standard error structure. With an appropriate choice of the resampling scheme, xtbcfe can take into account several heteroscedasticity and cross-sectional dependence... (More)
xtbcfe is a bootstrap-corrected fixed effects (LSDV) estimator for dynamic panel data models of general order. It estimates the specified model with the fixed effects estimator and corrects its small T bias (see Nickell, 1981) using a simplified but extended version of the approach presented in Everaert and Pozzi (2007). The algorithm evaluates the bias of the fixed effects estimator in a numerical way to avoid the use of analytical correction formulas. xtbcfe is therefore also applicable to higher order (more than one lag of the dependent variable) models with a potential non-standard error structure. With an appropriate choice of the resampling scheme, xtbcfe can take into account several heteroscedasticity and cross-sectional dependence patterns that would invalidate the standard (analytical) correction methods. Inference can be performed using either parametric or non-parametric bootstrap. (Less)
Please use this url to cite or link to this publication:
author
; and
publishing date
type
Non-textual form
publication status
published
subject
keywords
XTBCFE, STATA
publisher
Boston College Department of Economics Statistical Software Components
language
English
LU publication?
no
additional info
Software component
id
aed7faf9-3164-4976-93ec-82ea3cbb1f7a
date added to LUP
2018-11-08 15:21:04
date last changed
2018-11-21 21:43:03
@misc{aed7faf9-3164-4976-93ec-82ea3cbb1f7a,
  abstract     = {{xtbcfe is a bootstrap-corrected fixed effects (LSDV) estimator for dynamic panel data models of general order. It estimates the specified model with the fixed effects estimator and corrects its small T bias (see Nickell, 1981) using a simplified but extended version of the approach presented in Everaert and Pozzi (2007). The algorithm evaluates the bias of the fixed effects estimator in a numerical way to avoid the use of analytical correction formulas. xtbcfe is therefore also applicable to higher order (more than one lag of the dependent variable) models with a potential non-standard error structure. With an appropriate choice of the resampling scheme, xtbcfe can take into account several heteroscedasticity and cross-sectional dependence patterns that would invalidate the standard (analytical) correction methods. Inference can be performed using either parametric or non-parametric bootstrap.}},
  author       = {{De Vos, Ignace and Everaert, Gerdie and Ruyssen, Ilse}},
  keywords     = {{XTBCFE; STATA}},
  language     = {{eng}},
  publisher    = {{Boston College Department of Economics Statistical Software Components}},
  title        = {{XTBCFE: Stata module to perform bootstrap-corrected Fixed Effects estimation and inference in dynamic panel models}},
  year         = {{2015}},
}