Gary Koop, Dale Poirier and Justin L. Tobias, "Semiparametric Bayesian
Inference in Multiple Equation Models", Journal of Applied Econometrics,
Vol. 20, No. 6, 2005, pp. 723-747.

This data set consists of N=303 observations on white males in 1992 taken
from the National Longitudinal Survey of Youth. 

The data set contains 10 variables. See paper for further details.

  Column 1:  Log hourly wage
  Column 2:  Education (years of schooling)
  Column 3:  Ability based on standardized AFQT test score
  Column 4:  Education of oldest sibling (years of schooling)
  Column 5:  Experience (total weeks of labor market experience)
  Column 6:  Tenure (weeks on current job)
  Column 7:  Mother's education (years of schooling)
  Column 8:  Father's education (years of schooling)
  Column 9:  Dummy variable for urban residence
  Column 10: Broken home dummy (dummy for living with both parents at age 14)

(NOTE: The paper also uses a continuous local unemployment rate variable,
but this variable is not included here because it is part of the restricted
GEOCODE version of the NLSY). 

The data are in the file kpt.dat, which is an ASCII file in DOS format. It
is zipped in kpt-data.zip. Unix users should use "unzip -a".