Estimation in Binary Choice Models with Measurement Errors

Research output: Working paper/PreprintWorking paper

98 Downloads (Pure)


In this paper we develop a simple maximum likelihood estimator for probit models where the regressors have measurement error. We first assume precise information about the reliability ratios (or, equivalently, the proxy correlations) of the regressors. We then show how reasonable bounds for the parameter estimates can be obtained when only imprecise information is available. The analysis is also extended to situations where the measurement error has non-zero mean and is correlated with the true values of the regressors. An extensive simulation study shows that the estimator works very well, even in quite small samples. Finally the method is applied to data explaining sick leave in Sweden
Original languageEnglish
PublisherDepartment of Economics, Lund University
Publication statusUnpublished - 2003

Publication series

NameWorking Papers, Department of Economics, Lund University

Subject classification (UKÄ)

  • Economics

Free keywords

  • Measurement error
  • errors-in-variables
  • probit
  • binary choice
  • bounds


Dive into the research topics of 'Estimation in Binary Choice Models with Measurement Errors'. Together they form a unique fingerprint.

Cite this