DSpace
 

Osaka Prefecture University Repository >
D.Former Osaka Prefecture University:(1955-2004) >
0300.College of Economics/Graduate School of Economics >
Bulletin of Osaka Prefecture University. Ser. D, Economics, business administration and law >
Volume XLII >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10466/981

Title: <Article>Statistical Inference of Two Error Components Models In the Presence of Unknown Heteroskedasticity
Authors: Yoshida, Atsushi
Issue Date: 31-Mar-1998
Publisher: Osaka Prefecture University
Citation: Bulletin of Osaka Prefecture University. Ser. D, Economics, business administration and law. 1998, 42, p.45-64
Abstract: The purpose of this paper is to suggest an estimator which is more efficient than the within-class estimator for two error components models (2ECM) when the individual-specific error term is independent but not identically destributed. We will apply the idea of Amemiya's partially GLS (PGLS) to the model, which is regarded as an instrumental variables method using the regressors and the functions pf regressors as the instrumental variables. A specification test of the heteroskedasticity is also proposed.
URI: http://hdl.handle.net/10466/981
Appears in Collections:Volume XLII

Files in This Item:

File Description SizeFormat
KJ00000041536.pdf766.54 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.