Comparing features of convenient estimators for binary choice models with endogenous regressors

Lewbel, Arthur; Dong, Yingying; Yang, Thomas Tao

doi:10.1111/j.1540-5982.2012.01733.x

“…First, we estimate a linear probability model (LPM) because linear models do not suffer much from incidental parameter bias (Greene, 2004). Furthermore Lewbel et al (2012) …”

Section: Econometric Estimationmentioning

confidence: 99%

Social Network Effects on Mobile Money Adoption in Uganda

Murendo

¹

,

Wollni

²

,

Brauw

³

et al. 2017

The Journal of Development Studies

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. -551-3920212, fax: +49-551-3920203, e-mail: conrad.murendo@agr.uni-goettingen.de Abstract Social networks play a vital role in generating social learning and information exchange that can drive the diffusion of new financial innovations. This is particularly relevant for developing countries where education, extension and financial information services are underprovided. The recent introduction of mobile money in Africa represents a case where imperfect financial markets, weak extension services and information asymmetries limit the ability of rural households to make informed decisions to take advantage of mobile money innovation. This article identifies the role of social networks in the adoption of mobile money in Uganda. Using data from a survey of 477 rural households, a probit model is estimated controlling for household characteristics, correlated effects, and other possible information sources. Results suggest that learning within social networks helps disseminate information about mobile money and has enhanced its adoption. Compared to poor households, non-poor households rely more on social networks for information about mobile money. Mobile money adoption is likely to be enhanced if promotion programs reach more social networks. Terms of use: Documents in

show abstract

“…Hence, they are 'true' instrumental variable estimators (Lewbel et al, 2012). Yet linear models also have a number of less convenient features when they are applied to dichotomous variables.…”

Section: Resultsmentioning

confidence: 99%

Hard to Forget: The Long-Lasting Impact of War on Mental Health

Bratti

¹

,

Mendola

²

,

Miranda

³

2015

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in D I S C U S S I O N P A P E R S E R I E SHard The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center and a place of communication between science, politics and business. IZA is an independent nonprofit organization supported by Deutsche Post Foundation. The center is associated with the University of Bonn and offers a stimulating research environment through its international network, workshops and conferences, data service, project support, research visits and doctoral program. IZA engages in (i) original and internationally competitive research in all fields of labor economics, (ii) development of policy concepts, and (iii) dissemination of research results and concepts to the interested public.IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper should account for its provisional character. A revised version may be available directly from the author. ABSTRACTHard to Forget: The Long-Lasting Impact of War on Mental Health * War can have long-lasting effects on individual mental health through war trauma. In this paper, we explore the impact of constantly recalling painful episodes related to the 1992-1995 Bosnia and Herzegovina conflict on individual mental health in 2001 using the Center for Epidemiologic Studies Depression (CES-D) scale. Potential endogeneity and reverse causality issues are addressed using objective measures of war intensity recorded at the municipality level. We find that individuals experiencing war trauma have worse mental health six years after the end of the conflict. In particular, instrumental-variable estimates show that they score 16 points (more than 1.5 standard deviations) higher on the CES-D scale (with higher scores meaning more depression symptoms) and have a 60 percentage points higher probability of being at risk of depression. Our results are robust to a number of sensitivity checks accounting for individual geographical mobility and different treatment intensities, and suggest that the negative effects on mental health are not mainly mediated by physical health problems. Back-of-the envelope calculations show large economic costs of war-trauma. JEL Classification:I1, O1

show abstract

“…Papers discussing limiting distribution theory for special regressor based estimators include Lewbel (2000Lewbel ( , 2007a, Maurin (2007, 2008), Jacho-Chávez (2009), Khan and Tamer (2010), Nekipelov (2010a, 2010b), and Dong and Lewbel (2012). Lewbel, Dong, and Yang (2012) provide a comparison of special regressor models versus maximum likelihood estimation, control function estimators and linear probability models. They conclude that the greatest weakness of special regressor methods is the extent to which they rely on strong properties of just one regressor, some of which are difficult to verify, and the resulting sensitivity of estimates to this one regressor.…”

Section: Discussionmentioning

confidence: 99%

“…This is particularly useful for cases where some of the endogenous regressors are themselves discrete or limited. See Lewbel, Dong, and Yang (2012) for details.…”

Section: Regressorsmentioning

confidence: 99%

“…Other related surveys include Dong and Lewbel (2012), who describe the simplest estimators for applying special regressor methods to binary choice problems (particularly those involving endogenous regressors), and Lewbel, Dong, and Yang (2012) who provide a comparison of special regressor methods I would like to thank Thomas (Tao) Yang for helpful research assistance. Corresponding address: Arthur Lewbel, Dept of Economics, Boston College, 140 Commonwealth Ave., Chestnut Hill, MA, 02467, USA.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation