Instrumental Variable Estimation of Dynamic Linear Panel Data Models with Defactored Regressors and a Multifactor Error Structure

Norkutė, Milda; Sarafidis, Vasilis; Yamagata, Takashi; Cui, Guowei

Instrumental Variable Estimation of Dynamic Linear Panel Data Models with Defactored Regressors and a Multifactor Error Structure

Norkutė, Milda; Sarafidis, Vasilis; Yamagata, Takashi et al.

Institute of Social and Economic Research Discussion Papers, 2019, 1019, 1-C.12

アクセス数:157件（2025-04-30 03:14 集計）

固定URL: https://hdl.handle.net/11094/67810

論文情報

ファイル出力 EndNote Basic出力 Mendeley出力

タイトル	Instrumental Variable Estimation of Dynamic Linear Panel Data Models with Defactored Regressors and a Multifactor Error Structure
著者	著者名 Norkutė, Milda
	著者名 Sarafidis, Vasilis
	NRID 1000020813231 1000020813231 researchmap permalink takashi_yamagata takashi_yamagata Scopus Author ID 10138838100 10138838100 著者名 Yamagata, Takashi
	著者名 Cui, Guowei
キーワード等	method of moments
	dynamic panel data
	cross-sectional dependence
	factor model
内容	February 2018. Revised April 2019.
抄録	This paper develops two instrumental variable (IV) estimators for dynamic panel data models with exogenous covariates and a multifactor error structure when both cross-sectional and time series dimensions, N and T respectively, are large. Our approach initially projects out the common factors from the exogenous covariates of the model, and constructs instruments based on this defactored covariates. For models with homogeneous slope coefficients, we propose a two-step IV estimator: the first step IV estimator is obtained using the defactored covariates as instruments. In the second step, the entire model is defactored by the extracted factors from the residuals of the first step estimation and subsequently obtain the final IV estimator. For models with heterogeneous slope coefficients, we propose a mean-group type estimator, which is the cross-sectional average of first-step IV estimators of cross-section specific slopes. It is noteworthy that our estimators do not require us to seek for instrumental variables outside the model. Furthermore, our estimators are linear hence computationally robust and inexpensive. Moreover, they require no bias correction, and they are not subject to the small sample bias of least squares type estimators. The finite sample performances of the proposed estimators and associated statistical tests are investigated, and the results show that the estimators and the tests perform well even for small N and T.
出版者	The Institute of Social and Economic Research, Osaka University
収録物名	Institute of Social and Economic Research Discussion Papers
巻(号)	1019
ページ	1-C.12
刊行年月	2019-04
言語	英語
ハンドルURL	https://hdl.handle.net/11094/67810
関連情報 (references)	URI https://www.iser.osaka-u.ac.jp/library/dp/2018/DP1019R.pdf
アクセス権	metadata only access
カテゴリ	ディスカッション／ワーキング・ペーパー DP/WP
カテゴリ	Institute of Social and Economic Research Discussion Papers
最終更新日	2019-04

資源タイプ	テクニカルレポート
ローカル資源タイプ	ディスカッション／ワーキング・ペーパー
DCMI資源タイプ	text
DCTERMS.bibliographicCitation	Institute of Social and Economic Research Discussion Papers. 2019, 1019, p. 1-C.12
DC.title	Instrumental Variable Estimation of Dynamic Linear Panel Data Models with Defactored Regressors and a Multifactor Error Structure
DC.creator	Norkutė, Milda
	Sarafidis, Vasilis
	Yamagata, Takashi
	Cui, Guowei
DC.publisher	The Institute of Social and Economic Research, Osaka University
DC.language' scheme='DCTERMS.RFC1766	英語
DC.type' scheme='DCTERMS.DCMIType	text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2019-04
DC.relation' scheme='DCTERMS.URI	https://www.iser.osaka-u.ac.jp/library/dp/2018/DP1019R.pdf
DC.subject	method of moments
	dynamic panel data
	cross-sectional dependence
	factor model
DC.description	February 2018. Revised April 2019.
DCTERMS.abstract	This paper develops two instrumental variable (IV) estimators for dynamic panel data models with exogenous covariates and a multifactor error structure when both cross-sectional and time series dimensions, N and T respectively, are large. Our approach initially projects out the common factors from the exogenous covariates of the model, and constructs instruments based on this defactored covariates. For models with homogeneous slope coefficients, we propose a two-step IV estimator: the first step IV estimator is obtained using the defactored covariates as instruments. In the second step, the entire model is defactored by the extracted factors from the residuals of the first step estimation and subsequently obtain the final IV estimator. For models with heterogeneous slope coefficients, we propose a mean-group type estimator, which is the cross-sectional average of first-step IV estimators of cross-section specific slopes. It is noteworthy that our estimators do not require us to seek for instrumental variables outside the model. Furthermore, our estimators are linear hence computationally robust and inexpensive. Moreover, they require no bias correction, and they are not subject to the small sample bias of least squares type estimators. The finite sample performances of the proposed estimators and associated statistical tests are investigated, and the results show that the estimators and the tests perform well even for small N and T.
DC.format	67810
citation_title	Instrumental Variable Estimation of Dynamic Linear Panel Data Models with Defactored Regressors and a Multifactor Error Structure
citation_author	Norkutė, Milda
	Sarafidis, Vasilis
	Yamagata, Takashi
	Cui, Guowei
citation_publisher	The Institute of Social and Economic Research, Osaka University
citation_language	英語
citation_date	2019-04
citation_journal_title	Institute of Social and Economic Research Discussion Papers
citation_volume	1019
citation_firstpage	1
citation_lastpage	C.12
citation_public_url	https://hdl.handle.net/11094/67810
citation_abstract_html_url	https://www.iser.osaka-u.ac.jp/library/dp/2018/DP1019R.pdf

Instrumental Variable Estimation of Dynamic Linear Panel Data Models with Defactored Regressors and a Multifactor Error Structure

論文情報

論文詳細を表示