On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data

Shihan Du, Pia Homrighausen, Ralf Wilke

Publikation: Kapitel i bog/rapport/konferenceprocesKonferencebidrag i proceedingsForskningpeer review

Resumé

In virtually any empirical regression analysis, there is limited availability of observed variables and limited prior knowledge on which variables belong to the model. This paper provides a unified framework that nests various approaches aiming at reducing omitted variable bias in linear regression analysis. We work out the mechanisms driving the size of the bias and how various models with different regressor sets or unobserved effects relate. Without imposing restrictions on the relationship and role of the variables, it is, however, not possible to derive model rankings that are valid in every application.
In our applications, we apply the various models to German linked administrative labour market data. We find evidence for sizeable omitted variable bias in a wage regression, while a labour market transition model appears to be less affected. Additional survey variables contribute only to the wage model, while the use of work history variables and panel models lead to changes in coefficients in the two models. Overall, panel data models with a restricted regressor set are found to control for more unobserved information than cross-sectional analysis with an extended variable set.
In virtually any empirical regression analysis, there is limited availability of observed variables and limited prior knowledge on which variables belong to the model. This paper provides a unified framework that nests various approaches aiming at reducing omitted variable bias in linear regression analysis. We work out the mechanisms driving the size of the bias and how various models with different regressor sets or unobserved effects relate. Without imposing restrictions on the relationship and role of the variables, it is, however, not possible to derive model rankings that are valid in every application.
In our applications, we apply the various models to German linked administrative labour market data. We find evidence for sizeable omitted variable bias in a wage regression, while a labour market transition model appears to be less affected. Additional survey variables contribute only to the wage model, while the use of work history variables and panel models lead to changes in coefficients in the two models. Overall, panel data models with a restricted regressor set are found to control for more unobserved information than cross-sectional analysis with an extended variable set.
SprogEngelsk
TitelSymposium i anvendt statistik : 28.-30. januar 2019
Antal sider1
Udgivelses stedKøbenhavn
ForlagKøbenhavns Universitet
Dato2019
Sider1
ISBN (Trykt)9788779043596
StatusUdgivet - 2019
Begivenhed41. Symposium i Anvendt Statistik - Københavns Universitet, København, Danmark
Varighed: 28 jan. 201930 jan. 2019
http://statistiksymposium.dk/
http://www.statistiksymposium.dk/

Konference

Konference41. Symposium i Anvendt Statistik
LokationKøbenhavns Universitet
LandDanmark
ByKøbenhavn
Periode28/01/201930/01/2019
Internetadresse

Citer dette

Du, S., Homrighausen, P., & Wilke, R. (2019). On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data. I Symposium i anvendt statistik: 28.-30. januar 2019 (s. 1). København: Københavns Universitet.
Du, Shihan ; Homrighausen, Pia ; Wilke, Ralf. / On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data. Symposium i anvendt statistik: 28.-30. januar 2019. København : Københavns Universitet, 2019. s. 1
@inproceedings{154af03ee5714c69832e6baeb052abe2,
title = "On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data",
abstract = "In virtually any empirical regression analysis, there is limited availability of observed variables and limited prior knowledge on which variables belong to the model. This paper provides a unified framework that nests various approaches aiming at reducing omitted variable bias in linear regression analysis. We work out the mechanisms driving the size of the bias and how various models with different regressor sets or unobserved effects relate. Without imposing restrictions on the relationship and role of the variables, it is, however, not possible to derive model rankings that are valid in every application.In our applications, we apply the various models to German linked administrative labour market data. We find evidence for sizeable omitted variable bias in a wage regression, while a labour market transition model appears to be less affected. Additional survey variables contribute only to the wage model, while the use of work history variables and panel models lead to changes in coefficients in the two models. Overall, panel data models with a restricted regressor set are found to control for more unobserved information than cross-sectional analysis with an extended variable set.",
author = "Shihan Du and Pia Homrighausen and Ralf Wilke",
year = "2019",
language = "English",
isbn = "9788779043596",
pages = "1",
booktitle = "Symposium i anvendt statistik",
publisher = "K{\o}benhavns Universitet",
address = "Denmark",

}

Du, S, Homrighausen, P & Wilke, R 2019, On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data. i Symposium i anvendt statistik: 28.-30. januar 2019. Københavns Universitet, København, s. 1, 41. Symposium i Anvendt Statistik, København, Danmark, 28/01/2019.

On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data. / Du, Shihan; Homrighausen, Pia; Wilke, Ralf.

Symposium i anvendt statistik: 28.-30. januar 2019. København : Københavns Universitet, 2019. s. 1.

Publikation: Kapitel i bog/rapport/konferenceprocesKonferencebidrag i proceedingsForskningpeer review

TY - GEN

T1 - On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data

AU - Du,Shihan

AU - Homrighausen,Pia

AU - Wilke,Ralf

PY - 2019

Y1 - 2019

N2 - In virtually any empirical regression analysis, there is limited availability of observed variables and limited prior knowledge on which variables belong to the model. This paper provides a unified framework that nests various approaches aiming at reducing omitted variable bias in linear regression analysis. We work out the mechanisms driving the size of the bias and how various models with different regressor sets or unobserved effects relate. Without imposing restrictions on the relationship and role of the variables, it is, however, not possible to derive model rankings that are valid in every application.In our applications, we apply the various models to German linked administrative labour market data. We find evidence for sizeable omitted variable bias in a wage regression, while a labour market transition model appears to be less affected. Additional survey variables contribute only to the wage model, while the use of work history variables and panel models lead to changes in coefficients in the two models. Overall, panel data models with a restricted regressor set are found to control for more unobserved information than cross-sectional analysis with an extended variable set.

AB - In virtually any empirical regression analysis, there is limited availability of observed variables and limited prior knowledge on which variables belong to the model. This paper provides a unified framework that nests various approaches aiming at reducing omitted variable bias in linear regression analysis. We work out the mechanisms driving the size of the bias and how various models with different regressor sets or unobserved effects relate. Without imposing restrictions on the relationship and role of the variables, it is, however, not possible to derive model rankings that are valid in every application.In our applications, we apply the various models to German linked administrative labour market data. We find evidence for sizeable omitted variable bias in a wage regression, while a labour market transition model appears to be less affected. Additional survey variables contribute only to the wage model, while the use of work history variables and panel models lead to changes in coefficients in the two models. Overall, panel data models with a restricted regressor set are found to control for more unobserved information than cross-sectional analysis with an extended variable set.

UR - https://primo.kb.dk/primo-explore/fulldisplay?docid=CBS01000958648&context=L&vid=CBS&search_scope=Blended&isFrbr=true&tab=default_tab&lang=en_US

M3 - Article in proceedings

SN - 9788779043596

SP - 1

BT - Symposium i anvendt statistik

PB - Københavns Universitet

CY - København

ER -

Du S, Homrighausen P, Wilke R. On Omitted Variables, Proxies and Unobserved Effects in Analysis of Administrative Labour Market Data. I Symposium i anvendt statistik: 28.-30. januar 2019. København: Københavns Universitet. 2019. s. 1.