RT Journal Article
T1 Symmetrically normalized instrumental-variable estimation using panel data
A1 Alonso-Borrego, César
A1 Arellano, Manuel
AB We discuss the estimation of linear panel-data models with sequential moment restrictionsu sing symmetricallyn or malizedg eneralized method of moments( GMM) estimators( SNM)and limited information maximuml i kelihood( LIML)analogues These es imators are asymptotically equivalent to standardG MMb ut are invariantto normalizationan dt end to havea smallerf inite-samplbe ias, especiallyw hen the instruments are poor. We study their properties in relation to ordinary GMM and minimum distancee stimators for AR(1)model swith individual effects by mean of simulations. Finally, as empirical ilustrations, we estimate by SNM and LML employment and wage equations using panels of U.K. and Spanish firm.
PB American Statistical Association
SN 0735-0015
YR 1999
FD 1999-01
LK http://hdl.handle.net/10016/4655
UL http://hdl.handle.net/10016/4655
LA eng
NO The first author acknowledges research funding from a CEMFIP h.D. scholarship and from the SpanishDGES, Grant PB95-0292
DS e-Archivo
RD 28 abr. 2024