"Match Bias From Earnings Imputation in the Current Population Survey: The Case of Imperfect Matching"
at the Andrew Young School of Policy Studies

The Applied Econometrics Workshop for the spring semester continues on Friday, April 22, 2005, at 3:00 p.m. in the Smart Seminar Room 750 (7th floor of the Andrew Young School of Policy Studies Building, 14 Marietta Street, NW).

The speaker will be Chris R. Bollinger (right) of the University of Kentucky.

He will present his work on "Match Bias From Earnings Imputation in the Current Population Survey: The Case of Imperfect Matching." His paper examines alternative forms of match bias arising from earnings imputation. Wage equation parameters are estimated based on mixed samples of workers who report earnings and those who do not, the latter group being assigned earnings of donors who share some but not all the attributes of the recipients. Prior research demonstrates that regressions that include attributes not used as imputation match criteria are severely biased. The paper provides a general analytic expression to correct match bias in regression coefficients. This working paper (with Barry Hirsch of Trinity University) can be accessed on the intranet here, and more information about Dr. Bollinger can be found on his homepage.

Please contact Shif Gurmu at sgurmu@gsu.edu or (404) 651-1907 for more information.